pandoc.hs · typeable/pandoc

git.iko.soy/typeable/pandoc

on sable help sign in

typeable/pandoc/pandoc.hs

1171 lines 51 kB history raw

1{-# LANGUAGE CPP #-}

2{-

5This program is free software; you can redistribute it and/or modify

6it under the terms of the GNU General Public License as published by

7the Free Software Foundation; either version 2 of the License, or

8(at your option) any later version.

10This program is distributed in the hope that it will be useful,

11but WITHOUT ANY WARRANTY; without even the implied warranty of

12MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the

13GNU General Public License for more details.

15You should have received a copy of the GNU General Public License

16along with this program; if not, write to the Free Software

17Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA

18-}

20{- |

21 Module : Main

23 License : GNU GPL, version 2 or above

25 Maintainer : John MacFarlane <jgm@berkeley@edu>

26 Stability : alpha

27 Portability : portable

29Parses command-line options and calls the appropriate readers and

30writers.

31-}

32module Main where

33import Text.Pandoc

34import Text.Pandoc.Builder (setMeta)

35import Text.Pandoc.PDF (makePDF)

36import Text.Pandoc.Readers.LaTeX (handleIncludes)

37import Text.Pandoc.Shared ( tabFilter, readDataFileUTF8, readDataFile,

38 safeRead, headerShift, normalize, err, warn )

39import Text.Pandoc.XML ( toEntities )

40import Text.Pandoc.SelfContained ( makeSelfContained )

41import Text.Pandoc.Process (pipeProcess)

42import Text.Highlighting.Kate ( languages, Style, tango, pygments,

43 espresso, zenburn, kate, haddock, monochrome )

44import System.Environment ( getArgs, getProgName )

45import System.Exit ( exitWith, ExitCode (..) )

46import System.FilePath

47import System.Console.GetOpt

48import Data.Char ( toLower )

49import Data.List ( intercalate, isPrefixOf, sort )

50import System.Directory ( getAppUserDataDirectory, findExecutable )

51import System.IO ( stdout, stderr )

52import System.IO.Error ( isDoesNotExistError )

53import qualified Control.Exception as E

54import Control.Exception.Extensible ( throwIO )

55import qualified Text.Pandoc.UTF8 as UTF8

56import Control.Monad (when, unless, liftM)

57import Data.Foldable (foldrM)

58import Network.HTTP (simpleHTTP, mkRequest, getResponseBody, RequestMethod(..))

59import Network.URI (parseURI, isURI, URI(..))

60import qualified Data.ByteString.Lazy as B

61import qualified Data.ByteString as BS

62import Data.Aeson (eitherDecode', encode)

63import qualified Data.Map as M

64import System.IO.Error(ioeGetErrorType)

65import GHC.IO.Exception (IOErrorType(ResourceVanished))

66import Data.Yaml (decode)

67import qualified Data.Yaml as Yaml

68import qualified Data.Text as T

70copyrightMessage :: String

72 "Web: http://johnmacfarlane.net/pandoc\n" ++

73 "This is free software; see the source for copying conditions. There is no\n" ++

74 "warranty, not even for merchantability or fitness for a particular purpose."

76compileInfo :: String

77compileInfo =

78 "\nCompiled with texmath " ++

79 VERSION_texmath ++ ", highlighting-kate " ++ VERSION_highlighting_kate ++

80 ".\nSyntax highlighting is supported for the following languages:\n " ++

81 wrapWords 4 78

82 [map toLower l | l <- languages, l /= "Alert" && l /= "Alert_indent"]

84-- | Converts a list of strings into a single string with the items printed as

85-- comma separated words in lines with a maximum line length.

86wrapWords :: Int -> Int -> [String] -> String

87wrapWords indent c = wrap' (c - indent) (c - indent)

88 where wrap' _ _ [] = ""

89 wrap' cols remaining (x:xs) = if remaining == cols

90 then x ++ wrap' cols (remaining - length x) xs

91 else if (length x + 1) > remaining

92 then ",\n" ++ replicate indent ' ' ++ x ++ wrap' cols (cols - length x) xs

93 else ", " ++ x ++ wrap' cols (remaining - (length x + 2)) xs

95isTextFormat :: String -> Bool

96isTextFormat s = takeWhile (`notElem` "+-") s `notElem` ["odt","docx","epub","epub3"]

98externalFilter :: FilePath -> [String] -> Pandoc -> IO Pandoc

99externalFilter f args' d = do

100 (exitcode, outbs, errbs) <- E.handle filterException $

101 pipeProcess Nothing f args' $ encode d

102 when (not $ B.null errbs) $ B.hPutStr stderr errbs

103 case exitcode of

104 ExitSuccess -> return $ either error id $ eitherDecode' outbs

105 ExitFailure _ -> err 83 $ "Error running filter " ++ f

106 where filterException :: E.SomeException -> IO a

107 filterException e = err 83 $ "Error running filter " ++ f ++ "\n" ++

108 if ioeGetErrorType `fmap` E.fromException e ==

109 Just ResourceVanished

110 then f ++ " not found in path"

111 else show e

112

113-- | Data structure for command line options.

114data Opt = Opt

115 { optTabStop :: Int -- ^ Number of spaces per tab

116 , optPreserveTabs :: Bool -- ^ Preserve tabs instead of converting to spaces

117 , optStandalone :: Bool -- ^ Include header, footer

118 , optReader :: String -- ^ Reader format

119 , optWriter :: String -- ^ Writer format

120 , optParseRaw :: Bool -- ^ Parse unconvertable HTML and TeX

121 , optTableOfContents :: Bool -- ^ Include table of contents

122 , optTransforms :: [Pandoc -> Pandoc] -- ^ Doc transforms to apply

123 , optTemplate :: Maybe FilePath -- ^ Custom template

124 , optVariables :: [(String,String)] -- ^ Template variables to set

125 , optMetadata :: M.Map String MetaValue -- ^ Metadata fields to set

126 , optOutputFile :: String -- ^ Name of output file

127 , optNumberSections :: Bool -- ^ Number sections in LaTeX

128 , optNumberOffset :: [Int] -- ^ Starting number for sections

129 , optSectionDivs :: Bool -- ^ Put sections in div tags in HTML

130 , optIncremental :: Bool -- ^ Use incremental lists in Slidy/Slideous/S5

131 , optSelfContained :: Bool -- ^ Make HTML accessible offline

132 , optSmart :: Bool -- ^ Use smart typography

133 , optOldDashes :: Bool -- ^ Parse dashes like pandoc <=1.8.2.1

134 , optHtml5 :: Bool -- ^ Produce HTML5 in HTML

135 , optHtmlQTags :: Bool -- ^ Use <q> tags in HTML

136 , optHighlight :: Bool -- ^ Highlight source code

137 , optHighlightStyle :: Style -- ^ Style to use for highlighted code

138 , optChapters :: Bool -- ^ Use chapter for top-level sects

139 , optHTMLMathMethod :: HTMLMathMethod -- ^ Method to print HTML math

140 , optReferenceODT :: Maybe FilePath -- ^ Path of reference.odt

141 , optReferenceDocx :: Maybe FilePath -- ^ Path of reference.docx

142 , optEpubStylesheet :: Maybe String -- ^ EPUB stylesheet

143 , optEpubMetadata :: String -- ^ EPUB metadata

144 , optEpubFonts :: [FilePath] -- ^ EPUB fonts to embed

145 , optEpubChapterLevel :: Int -- ^ Header level at which to split chapters

146 , optTOCDepth :: Int -- ^ Number of levels to include in TOC

147 , optDumpArgs :: Bool -- ^ Output command-line arguments

148 , optIgnoreArgs :: Bool -- ^ Ignore command-line arguments

149 , optReferenceLinks :: Bool -- ^ Use reference links in writing markdown, rst

150 , optWrapText :: Bool -- ^ Wrap text

151 , optColumns :: Int -- ^ Line length in characters

152 , optFilters :: [FilePath] -- ^ Filters to apply

153 , optEmailObfuscation :: ObfuscationMethod

154 , optIdentifierPrefix :: String

155 , optIndentedCodeClasses :: [String] -- ^ Default classes for indented code blocks

156 , optDataDir :: Maybe FilePath

157 , optCiteMethod :: CiteMethod -- ^ Method to output cites

158 , optListings :: Bool -- ^ Use listings package for code blocks

159 , optLaTeXEngine :: String -- ^ Program to use for latex -> pdf

160 , optSlideLevel :: Maybe Int -- ^ Header level that creates slides

161 , optSetextHeaders :: Bool -- ^ Use atx headers for markdown level 1-2

162 , optAscii :: Bool -- ^ Use ascii characters only in html

163 , optTeXLigatures :: Bool -- ^ Use TeX ligatures for quotes/dashes

164 , optDefaultImageExtension :: String -- ^ Default image extension

165 }

166

167-- | Defaults for command-line options.

168defaultOpts :: Opt

169defaultOpts = Opt

170 { optTabStop = 4

171 , optPreserveTabs = False

172 , optStandalone = False

173 , optReader = "" -- null for default reader

174 , optWriter = "" -- null for default writer

175 , optParseRaw = False

176 , optTableOfContents = False

177 , optTransforms = []

178 , optTemplate = Nothing

179 , optVariables = []

180 , optMetadata = M.empty

181 , optOutputFile = "-" -- "-" means stdout

182 , optNumberSections = False

183 , optNumberOffset = [0,0,0,0,0,0]

184 , optSectionDivs = False

185 , optIncremental = False

186 , optSelfContained = False

187 , optSmart = False

188 , optOldDashes = False

189 , optHtml5 = False

190 , optHtmlQTags = False

191 , optHighlight = True

192 , optHighlightStyle = pygments

193 , optChapters = False

194 , optHTMLMathMethod = PlainMath

195 , optReferenceODT = Nothing

196 , optReferenceDocx = Nothing

197 , optEpubStylesheet = Nothing

198 , optEpubMetadata = ""

199 , optEpubFonts = []

200 , optEpubChapterLevel = 1

201 , optTOCDepth = 3

202 , optDumpArgs = False

203 , optIgnoreArgs = False

204 , optReferenceLinks = False

205 , optWrapText = True

206 , optColumns = 72

207 , optFilters = []

208 , optEmailObfuscation = JavascriptObfuscation

209 , optIdentifierPrefix = ""

210 , optIndentedCodeClasses = []

211 , optDataDir = Nothing

212 , optCiteMethod = Citeproc

213 , optListings = False

214 , optLaTeXEngine = "pdflatex"

215 , optSlideLevel = Nothing

216 , optSetextHeaders = True

217 , optAscii = False

218 , optTeXLigatures = True

219 , optDefaultImageExtension = ""

220 }

221

222-- | A list of functions, each transforming the options data structure

223-- in response to a command-line option.

224options :: [OptDescr (Opt -> IO Opt)]

225options =

226 [ Option "fr" ["from","read"]

227 (ReqArg

228 (\arg opt -> return opt { optReader = map toLower arg })

229 "FORMAT")

230 ""

231

232 , Option "tw" ["to","write"]

233 (ReqArg

234 (\arg opt -> return opt { optWriter = map toLower arg })

235 "FORMAT")

236 ""

237

238 , Option "o" ["output"]

239 (ReqArg

240 (\arg opt -> return opt { optOutputFile = arg })

241 "FILENAME")

242 "" -- "Name of output file"

243

244 , Option "" ["data-dir"]

245 (ReqArg

246 (\arg opt -> return opt { optDataDir = Just arg })

247 "DIRECTORY") -- "Directory containing pandoc data files."

248 ""

249

250 , Option "" ["strict"]

251 (NoArg

252 (\opt -> do

253 err 59 $ "The --strict option has been removed.\n" ++

254 "Use `markdown_strict' input or output format instead."

255 return opt ))

256 "" -- "Disable markdown syntax extensions"

257

258 , Option "R" ["parse-raw"]

259 (NoArg

260 (\opt -> return opt { optParseRaw = True }))

261 "" -- "Parse untranslatable HTML codes and LaTeX environments as raw"

262

263 , Option "S" ["smart"]

264 (NoArg

265 (\opt -> return opt { optSmart = True }))

266 "" -- "Use smart quotes, dashes, and ellipses"

267

268 , Option "" ["old-dashes"]

269 (NoArg

270 (\opt -> return opt { optSmart = True

271 , optOldDashes = True }))

272 "" -- "Use smart quotes, dashes, and ellipses"

273

274 , Option "" ["base-header-level"]

275 (ReqArg

276 (\arg opt ->

277 case safeRead arg of

278 Just t | t > 0 -> do

279 let oldTransforms = optTransforms opt

280 let shift = t - 1

281 return opt{ optTransforms =

282 headerShift shift : oldTransforms }

283 _ -> err 19

284 "base-header-level must be a number > 0")

285 "NUMBER")

286 "" -- "Headers base level"

287

288 , Option "" ["indented-code-classes"]

289 (ReqArg

290 (\arg opt -> return opt { optIndentedCodeClasses = words $

291 map (\c -> if c == ',' then ' ' else c) arg })

292 "STRING")

293 "" -- "Classes (whitespace- or comma-separated) to use for indented code-blocks"

294

295 , Option "F" ["filter"]

296 (ReqArg

297 (\arg opt -> return opt { optFilters = arg : optFilters opt })

298 "PROGRAM")

299 "" -- "External JSON filter"

300

301 , Option "" ["normalize"]

302 (NoArg

303 (\opt -> return opt { optTransforms =

304 normalize : optTransforms opt } ))

305 "" -- "Normalize the Pandoc AST"

306

307 , Option "p" ["preserve-tabs"]

308 (NoArg

309 (\opt -> return opt { optPreserveTabs = True }))

310 "" -- "Preserve tabs instead of converting to spaces"

311

312 , Option "" ["tab-stop"]

313 (ReqArg

314 (\arg opt ->

315 case safeRead arg of

316 Just t | t > 0 -> return opt { optTabStop = t }

317 _ -> err 31

318 "tab-stop must be a number greater than 0")

319 "NUMBER")

320 "" -- "Tab stop (default 4)"

321

322 , Option "s" ["standalone"]

323 (NoArg

324 (\opt -> return opt { optStandalone = True }))

325 "" -- "Include needed header and footer on output"

326

327 , Option "" ["template"]

328 (ReqArg

329 (\arg opt -> do

330 return opt{ optTemplate = Just arg,

331 optStandalone = True })

332 "FILENAME")

333 "" -- "Use custom template"

334

335 , Option "M" ["metadata"]

336 (ReqArg

337 (\arg opt -> do

338 let (key,val) = case break (`elem` ":=") arg of

339 (k,_:v) -> (k, readMetaValue v)

340 (k,_) -> (k, MetaBool True)

341 return opt{ optMetadata = addMetadata key val

342 $ optMetadata opt })

343 "KEY[:VALUE]")

344 ""

345

346 , Option "V" ["variable"]

347 (ReqArg

348 (\arg opt -> do

349 let (key,val) = case break (`elem` ":=") arg of

350 (k,_:v) -> (k,v)

351 (k,_) -> (k,"true")

352 return opt{ optVariables = (key,val) : optVariables opt })

353 "KEY[:VALUE]")

354 ""

355

356 , Option "D" ["print-default-template"]

357 (ReqArg

358 (\arg _ -> do

359 templ <- getDefaultTemplate Nothing arg

360 case templ of

361 Right t -> UTF8.hPutStr stdout t

362 Left e -> error $ show e

363 exitWith ExitSuccess)

364 "FORMAT")

365 "" -- "Print default template for FORMAT"

366

367 , Option "" ["print-default-data-file"]

368 (ReqArg

369 (\arg _ -> do

370 readDataFile Nothing arg >>= BS.hPutStr stdout

371 exitWith ExitSuccess)

372 "FILE")

373 "" -- "Print default data file"

374

375 , Option "" ["no-wrap"]

376 (NoArg

377 (\opt -> return opt { optWrapText = False }))

378 "" -- "Do not wrap text in output"

379

380 , Option "" ["columns"]

381 (ReqArg

382 (\arg opt ->

383 case safeRead arg of

384 Just t | t > 0 -> return opt { optColumns = t }

385 _ -> err 33 $

386 "columns must be a number greater than 0")

387 "NUMBER")

388 "" -- "Length of line in characters"

389

390 , Option "" ["toc", "table-of-contents"]

391 (NoArg

392 (\opt -> return opt { optTableOfContents = True }))

393 "" -- "Include table of contents"

394

395 , Option "" ["toc-depth"]

396 (ReqArg

397 (\arg opt -> do

398 case safeRead arg of

399 Just t | t >= 1 && t <= 6 ->

400 return opt { optTOCDepth = t }

401 _ -> err 57 $

402 "TOC level must be a number between 1 and 6")

403 "NUMBER")

404 "" -- "Number of levels to include in TOC"

405

406 , Option "" ["no-highlight"]

407 (NoArg

408 (\opt -> return opt { optHighlight = False }))

409 "" -- "Don't highlight source code"

410

411 , Option "" ["highlight-style"]

412 (ReqArg

413 (\arg opt -> do

414 newStyle <- case map toLower arg of

415 "pygments" -> return pygments

416 "tango" -> return tango

417 "espresso" -> return espresso

418 "zenburn" -> return zenburn

419 "kate" -> return kate

420 "monochrome" -> return monochrome

421 "haddock" -> return haddock

422 _ -> err 39 $

423 "Unknown style :" ++ arg

424 return opt{ optHighlightStyle = newStyle })

425 "STYLE")

426 "" -- "Style for highlighted code"

427

428 , Option "H" ["include-in-header"]

429 (ReqArg

430 (\arg opt -> do

431 text <- UTF8.readFile arg

432 -- add new ones to end, so they're included in order specified

433 let newvars = optVariables opt ++ [("header-includes",text)]

434 return opt { optVariables = newvars,

435 optStandalone = True })

436 "FILENAME")

437 "" -- "File to include at end of header (implies -s)"

438

439 , Option "B" ["include-before-body"]

440 (ReqArg

441 (\arg opt -> do

442 text <- UTF8.readFile arg

443 -- add new ones to end, so they're included in order specified

444 let newvars = optVariables opt ++ [("include-before",text)]

445 return opt { optVariables = newvars,

446 optStandalone = True })

447 "FILENAME")

448 "" -- "File to include before document body"

449

450 , Option "A" ["include-after-body"]

451 (ReqArg

452 (\arg opt -> do

453 text <- UTF8.readFile arg

454 -- add new ones to end, so they're included in order specified

455 let newvars = optVariables opt ++ [("include-after",text)]

456 return opt { optVariables = newvars,

457 optStandalone = True })

458 "FILENAME")

459 "" -- "File to include after document body"

460

461 , Option "" ["self-contained"]

462 (NoArg

463 (\opt -> return opt { optSelfContained = True,

464 optVariables = ("slidy-url","slidy") :

465 optVariables opt,

466 optStandalone = True }))

467 "" -- "Make slide shows include all the needed js and css"

468

469 , Option "" ["offline"]

470 (NoArg

471 (\opt -> do warn $ "--offline is deprecated. Use --self-contained instead."

472 return opt { optSelfContained = True,

473 optStandalone = True }))

474 "" -- "Make slide shows include all the needed js and css"

475 -- deprecated synonym for --self-contained

476

477 , Option "5" ["html5"]

478 (NoArg

479 (\opt -> do

480 warn $ "--html5 is deprecated. "

481 ++ "Use the html5 output format instead."

482 return opt { optHtml5 = True }))

483 "" -- "Produce HTML5 in HTML output"

484

485 , Option "" ["html-q-tags"]

486 (NoArg

487 (\opt -> do

488 return opt { optHtmlQTags = True }))

489 "" -- "Use <q> tags for quotes in HTML"

490

491 , Option "" ["ascii"]

492 (NoArg

493 (\opt -> return opt { optAscii = True }))

494 "" -- "Use ascii characters only in HTML output"

495

496 , Option "" ["reference-links"]

497 (NoArg

498 (\opt -> return opt { optReferenceLinks = True } ))

499 "" -- "Use reference links in parsing HTML"

500

501 , Option "" ["atx-headers"]

502 (NoArg

503 (\opt -> return opt { optSetextHeaders = False } ))

504 "" -- "Use atx-style headers for markdown"

505

506 , Option "" ["chapters"]

507 (NoArg

508 (\opt -> return opt { optChapters = True }))

509 "" -- "Use chapter for top-level sections in LaTeX, DocBook"

510

511 , Option "N" ["number-sections"]

512 (NoArg

513 (\opt -> return opt { optNumberSections = True }))

514 "" -- "Number sections in LaTeX"

515

516 , Option "" ["number-offset"]

517 (ReqArg

518 (\arg opt ->

519 case safeRead ('[':arg ++ "]") of

520 Just ns -> return opt { optNumberOffset = ns,

521 optNumberSections = True }

522 _ -> err 57 "could not parse number-offset")

523 "NUMBERS")

524 "" -- "Starting number for sections, subsections, etc."

525

526 , Option "" ["no-tex-ligatures"]

527 (NoArg

528 (\opt -> return opt { optTeXLigatures = False }))

529 "" -- "Don't use tex ligatures for quotes, dashes"

530

531 , Option "" ["listings"]

532 (NoArg

533 (\opt -> return opt { optListings = True }))

534 "" -- "Use listings package for LaTeX code blocks"

535

536 , Option "i" ["incremental"]

537 (NoArg

538 (\opt -> return opt { optIncremental = True }))

539 "" -- "Make list items display incrementally in Slidy/Slideous/S5"

540

541 , Option "" ["slide-level"]

542 (ReqArg

543 (\arg opt -> do

544 case safeRead arg of

545 Just t | t >= 1 && t <= 6 ->

546 return opt { optSlideLevel = Just t }

547 _ -> err 39 $

548 "slide level must be a number between 1 and 6")

549 "NUMBER")

550 "" -- "Force header level for slides"

551

552 , Option "" ["section-divs"]

553 (NoArg

554 (\opt -> return opt { optSectionDivs = True }))

555 "" -- "Put sections in div tags in HTML"

556

557 , Option "" ["default-image-extension"]

558 (ReqArg

559 (\arg opt -> return opt { optDefaultImageExtension = arg })

560 "extension")

561 "" -- "Default extension for extensionless images"

562

563 , Option "" ["email-obfuscation"]

564 (ReqArg

565 (\arg opt -> do

566 method <- case arg of

567 "references" -> return ReferenceObfuscation

568 "javascript" -> return JavascriptObfuscation

569 "none" -> return NoObfuscation

570 _ -> err 6

571 ("Unknown obfuscation method: " ++ arg)

572 return opt { optEmailObfuscation = method })

573 "none|javascript|references")

574 "" -- "Method for obfuscating email in HTML"

575

576 , Option "" ["id-prefix"]

577 (ReqArg

578 (\arg opt -> return opt { optIdentifierPrefix = arg })

579 "STRING")

580 "" -- "Prefix to add to automatically generated HTML identifiers"

581

582 , Option "T" ["title-prefix"]

583 (ReqArg

584 (\arg opt -> do

585 let newvars = ("title-prefix", arg) : optVariables opt

586 return opt { optVariables = newvars,

587 optStandalone = True })

588 "STRING")

589 "" -- "String to prefix to HTML window title"

590

591 , Option "c" ["css"]

592 (ReqArg

593 (\arg opt -> do

594 -- add new link to end, so it is included in proper order

595 let newvars = optVariables opt ++ [("css",arg)]

596 return opt { optVariables = newvars,

597 optStandalone = True })

598 "URL")

599 "" -- "Link to CSS style sheet"

600

601 , Option "" ["reference-odt"]

602 (ReqArg

603 (\arg opt -> do

604 return opt { optReferenceODT = Just arg })

605 "FILENAME")

606 "" -- "Path of custom reference.odt"

607

608 , Option "" ["reference-docx"]

609 (ReqArg

610 (\arg opt -> do

611 return opt { optReferenceDocx = Just arg })

612 "FILENAME")

613 "" -- "Path of custom reference.docx"

614

615 , Option "" ["epub-stylesheet"]

616 (ReqArg

617 (\arg opt -> do

618 text <- UTF8.readFile arg

619 return opt { optEpubStylesheet = Just text })

620 "FILENAME")

621 "" -- "Path of epub.css"

622

623 , Option "" ["epub-cover-image"]

624 (ReqArg

625 (\arg opt ->

626 return opt { optVariables =

627 ("epub-cover-image", arg) : optVariables opt })

628 "FILENAME")

629 "" -- "Path of epub cover image"

630

631 , Option "" ["epub-metadata"]

632 (ReqArg

633 (\arg opt -> do

634 text <- UTF8.readFile arg

635 return opt { optEpubMetadata = text })

636 "FILENAME")

637 "" -- "Path of epub metadata file"

638

639 , Option "" ["epub-embed-font"]

640 (ReqArg

641 (\arg opt -> do

642 return opt{ optEpubFonts = arg : optEpubFonts opt })

643 "FILE")

644 "" -- "Directory of fonts to embed"

645

646 , Option "" ["epub-chapter-level"]

647 (ReqArg

648 (\arg opt -> do

649 case safeRead arg of

650 Just t | t >= 1 && t <= 6 ->

651 return opt { optEpubChapterLevel = t }

652 _ -> err 59 $

653 "chapter level must be a number between 1 and 6")

654 "NUMBER")

655 "" -- "Header level at which to split chapters in EPUB"

656

657 , Option "" ["latex-engine"]

658 (ReqArg

659 (\arg opt -> do

660 let b = takeBaseName arg

661 if (b == "pdflatex" || b == "lualatex" || b == "xelatex")

662 then return opt { optLaTeXEngine = arg }

663 else err 45 "latex-engine must be pdflatex, lualatex, or xelatex.")

664 "PROGRAM")

665 "" -- "Name of latex program to use in generating PDF"

666

667 , Option "" ["bibliography"]

668 (ReqArg

669 (\arg opt -> return opt{ optMetadata = addMetadata

670 "bibliography" (readMetaValue arg)

671 $ optMetadata opt

672 })

673 "FILE")

674 ""

675

676 , Option "" ["csl"]

677 (ReqArg

678 (\arg opt ->

679 return opt{ optMetadata = addMetadata "csl"

680 (readMetaValue arg)

681 $ optMetadata opt })

682 "FILE")

683 ""

684

685 , Option "" ["citation-abbreviations"]

686 (ReqArg

687 (\arg opt ->

688 return opt{ optMetadata = addMetadata

689 "citation-abbreviations"

690 (readMetaValue arg)

691 $ optMetadata opt })

692 "FILE")

693 ""

694

695 , Option "" ["natbib"]

696 (NoArg

697 (\opt -> return opt { optCiteMethod = Natbib }))

698 "" -- "Use natbib cite commands in LaTeX output"

699

700 , Option "" ["biblatex"]

701 (NoArg

702 (\opt -> return opt { optCiteMethod = Biblatex }))

703 "" -- "Use biblatex cite commands in LaTeX output"

704

705 , Option "m" ["latexmathml", "asciimathml"]

706 (OptArg

707 (\arg opt ->

708 return opt { optHTMLMathMethod = LaTeXMathML arg })

709 "URL")

710 "" -- "Use LaTeXMathML script in html output"

711

712 , Option "" ["mathml"]

713 (OptArg

714 (\arg opt ->

715 return opt { optHTMLMathMethod = MathML arg })

716 "URL")

717 "" -- "Use mathml for HTML math"

718

719 , Option "" ["mimetex"]

720 (OptArg

721 (\arg opt -> do

722 let url' = case arg of

723 Just u -> u ++ "?"

724 Nothing -> "/cgi-bin/mimetex.cgi?"

725 return opt { optHTMLMathMethod = WebTeX url' })

726 "URL")

727 "" -- "Use mimetex for HTML math"

728

729 , Option "" ["webtex"]

730 (OptArg

731 (\arg opt -> do

732 let url' = case arg of

733 Just u -> u

734 Nothing -> "http://chart.apis.google.com/chart?cht=tx&chl="

735 return opt { optHTMLMathMethod = WebTeX url' })

736 "URL")

737 "" -- "Use web service for HTML math"

738

739 , Option "" ["jsmath"]

740 (OptArg

741 (\arg opt -> return opt { optHTMLMathMethod = JsMath arg})

742 "URL")

743 "" -- "Use jsMath for HTML math"

744

745 , Option "" ["mathjax"]

746 (OptArg

747 (\arg opt -> do

748 let url' = case arg of

749 Just u -> u

750 Nothing -> "http://cdn.mathjax.org/mathjax/latest/MathJax.js?config=TeX-AMS-MML_HTMLorMML"

751 return opt { optHTMLMathMethod = MathJax url'})

752 "URL")

753 "" -- "Use MathJax for HTML math"

754

755 , Option "" ["gladtex"]

756 (NoArg

757 (\opt -> return opt { optHTMLMathMethod = GladTeX }))

758 "" -- "Use gladtex for HTML math"

759

760 , Option "" ["dump-args"]

761 (NoArg

762 (\opt -> return opt { optDumpArgs = True }))

763 "" -- "Print output filename and arguments to stdout."

764

765 , Option "" ["ignore-args"]

766 (NoArg

767 (\opt -> return opt { optIgnoreArgs = True }))

768 "" -- "Ignore command-line arguments."

769

770 , Option "v" ["version"]

771 (NoArg

772 (\_ -> do

773 prg <- getProgName

774 defaultDatadir <- getAppUserDataDirectory "pandoc"

775 UTF8.hPutStrLn stdout (prg ++ " " ++ pandocVersion ++

776 compileInfo ++ "\nDefault user data directory: " ++

777 defaultDatadir ++ copyrightMessage)

778 exitWith ExitSuccess ))

779 "" -- "Print version"

780

781 , Option "h" ["help"]

782 (NoArg

783 (\_ -> do

784 prg <- getProgName

785 UTF8.hPutStr stdout (usageMessage prg options)

786 exitWith ExitSuccess ))

787 "" -- "Show help"

788

789 ]

790

791addMetadata :: String -> MetaValue -> M.Map String MetaValue

792 -> M.Map String MetaValue

793addMetadata k v m = case M.lookup k m of

794 Nothing -> M.insert k v m

795 Just (MetaList xs) -> M.insert k

796 (MetaList (xs ++ [v])) m

797 Just x -> M.insert k (MetaList [v, x]) m

798

799readMetaValue :: String -> MetaValue

800readMetaValue s = case decode (UTF8.fromString s) of

801 Just (Yaml.String t) -> MetaString $ T.unpack t

802 Just (Yaml.Bool b) -> MetaBool b

803 _ -> MetaString s

804

805-- Returns usage message

806usageMessage :: String -> [OptDescr (Opt -> IO Opt)] -> String

807usageMessage programName = usageInfo

808 (programName ++ " [OPTIONS] [FILES]" ++ "\nInput formats: " ++

809 (wrapWords 16 78 $ readers'names) ++ "\nOutput formats: " ++

810 (wrapWords 16 78 $ writers'names) ++

811 '\n' : replicate 16 ' ' ++

812 "[*for pdf output, use latex or beamer and -o FILENAME.pdf]\nOptions:")

813 where

814 writers'names = sort $ "pdf*" : map fst writers

815 readers'names = sort $ map fst readers

816

817-- Determine default reader based on source file extensions

818defaultReaderName :: String -> [FilePath] -> String

819defaultReaderName fallback [] = fallback

820defaultReaderName fallback (x:xs) =

821 case takeExtension (map toLower x) of

822 ".xhtml" -> "html"

823 ".html" -> "html"

824 ".htm" -> "html"

825 ".tex" -> "latex"

826 ".latex" -> "latex"

827 ".ltx" -> "latex"

828 ".rst" -> "rst"

829 ".lhs" -> "markdown+lhs"

830 ".db" -> "docbook"

831 ".opml" -> "opml"

832 ".wiki" -> "mediawiki"

833 ".textile" -> "textile"

834 ".native" -> "native"

835 ".json" -> "json"

836 _ -> defaultReaderName fallback xs

837

838-- Returns True if extension of first source is .lhs

839lhsExtension :: [FilePath] -> Bool

840lhsExtension (x:_) = takeExtension x == ".lhs"

841lhsExtension _ = False

842

843-- Determine default writer based on output file extension

844defaultWriterName :: FilePath -> String

845defaultWriterName "-" = "html" -- no output file

846defaultWriterName x =

847 case takeExtension (map toLower x) of

848 "" -> "markdown" -- empty extension

849 ".tex" -> "latex"

850 ".latex" -> "latex"

851 ".ltx" -> "latex"

852 ".context" -> "context"

853 ".ctx" -> "context"

854 ".rtf" -> "rtf"

855 ".rst" -> "rst"

856 ".s5" -> "s5"

857 ".native" -> "native"

858 ".json" -> "json"

859 ".txt" -> "markdown"

860 ".text" -> "markdown"

861 ".md" -> "markdown"

862 ".markdown" -> "markdown"

863 ".textile" -> "textile"

864 ".lhs" -> "markdown+lhs"

865 ".texi" -> "texinfo"

866 ".texinfo" -> "texinfo"

867 ".db" -> "docbook"

868 ".odt" -> "odt"

869 ".docx" -> "docx"

870 ".epub" -> "epub"

871 ".org" -> "org"

872 ".asciidoc" -> "asciidoc"

873 ".pdf" -> "latex"

874 ".fb2" -> "fb2"

875 ".sable" -> "sable"

876 ".opml" -> "opml"

877 ['.',y] | y `elem` ['1'..'9'] -> "man"

878 _ -> "html"

879

880main :: IO ()

881main = do

882

883 rawArgs <- liftM (map UTF8.decodeArg) getArgs

884 prg <- getProgName

885 let compatMode = (prg == "hsmarkdown")

886

887 let (actions, args, errors) = if compatMode

888 then ([], rawArgs, [])

889 else getOpt Permute options rawArgs

890

891 unless (null errors) $

892 err 2 $ concat $ errors ++

893 ["Try " ++ prg ++ " --help for more information."]

894

895 let defaultOpts' = if compatMode

896 then defaultOpts { optReader = "markdown_strict"

897 , optWriter = "html"

898 , optEmailObfuscation =

899 ReferenceObfuscation }

900 else defaultOpts

901

902 -- thread option data structure through all supplied option actions

903 opts <- foldl (>>=) (return defaultOpts') actions

904

905 let Opt { optTabStop = tabStop

906 , optPreserveTabs = preserveTabs

907 , optStandalone = standalone

908 , optReader = readerName

909 , optWriter = writerName

910 , optParseRaw = parseRaw

911 , optVariables = variables

912 , optMetadata = metadata

913 , optTableOfContents = toc

914 , optTransforms = transforms

915 , optTemplate = templatePath

916 , optOutputFile = outputFile

917 , optNumberSections = numberSections

918 , optNumberOffset = numberFrom

919 , optSectionDivs = sectionDivs

920 , optIncremental = incremental

921 , optSelfContained = selfContained

922 , optSmart = smart

923 , optOldDashes = oldDashes

924 , optHtml5 = html5

925 , optHtmlQTags = htmlQTags

926 , optHighlight = highlight

927 , optHighlightStyle = highlightStyle

928 , optChapters = chapters

929 , optHTMLMathMethod = mathMethod

930 , optReferenceODT = referenceODT

931 , optReferenceDocx = referenceDocx

932 , optEpubStylesheet = epubStylesheet

933 , optEpubMetadata = epubMetadata

934 , optEpubFonts = epubFonts

935 , optEpubChapterLevel = epubChapterLevel

936 , optTOCDepth = epubTOCDepth

937 , optDumpArgs = dumpArgs

938 , optIgnoreArgs = ignoreArgs

939 , optReferenceLinks = referenceLinks

940 , optWrapText = wrap

941 , optColumns = columns

942 , optFilters = filters

943 , optEmailObfuscation = obfuscationMethod

944 , optIdentifierPrefix = idPrefix

945 , optIndentedCodeClasses = codeBlockClasses

946 , optDataDir = mbDataDir

947 , optCiteMethod = citeMethod

948 , optListings = listings

949 , optLaTeXEngine = latexEngine

950 , optSlideLevel = slideLevel

951 , optSetextHeaders = setextHeaders

952 , optAscii = ascii

953 , optTeXLigatures = texLigatures

954 , optDefaultImageExtension = defaultImageExtension

955 } = opts

956

957 when dumpArgs $

958 do UTF8.hPutStrLn stdout outputFile

959 mapM_ (\arg -> UTF8.hPutStrLn stdout arg) args

960 exitWith ExitSuccess

961

962 -- --bibliography implies -F pandoc-citeproc for backwards compatibility:

963 let filters' = case M.lookup "bibliography" metadata of

964 Just _ | all (\f -> takeBaseName f /= "pandoc-citeproc")

965 filters -> "pandoc-citeproc" : filters

966 _ -> filters

967 let plugins = map externalFilter filters'

968

969 let sources = if ignoreArgs then [] else args

970

971 datadir <- case mbDataDir of

972 Nothing -> E.catch

973 (liftM Just $ getAppUserDataDirectory "pandoc")

974 (\e -> let _ = (e :: E.SomeException)

975 in return Nothing)

976 Just _ -> return mbDataDir

977

978 -- assign reader and writer based on options and filenames

979 let readerName' = if null readerName

980 then let fallback = if any isURI sources

981 then "html"

982 else "markdown"

983 in defaultReaderName fallback sources

984 else readerName

985

986 let writerName' = if null writerName

987 then defaultWriterName outputFile

988 else writerName

989

990 let pdfOutput = map toLower (takeExtension outputFile) == ".pdf"

991

992 let laTeXOutput = "latex" `isPrefixOf` writerName' ||

993 "beamer" `isPrefixOf` writerName'

994

995 when pdfOutput $ do

996 -- make sure writer is latex or beamer

997 unless laTeXOutput $

998 err 47 $ "cannot produce pdf output with " ++ writerName' ++ " writer"

999 -- check for latex program

1000 mbLatex <- findExecutable latexEngine

1001 case mbLatex of

1002 Nothing -> err 41 $

1003 latexEngine ++ " not found. " ++

1004 latexEngine ++ " is needed for pdf output."

1005 Just _ -> return ()

1006

1007 reader <- case getReader readerName' of

1008 Right r -> return r

1009 Left e -> err 7 e

1010

1011 let standalone' = standalone || not (isTextFormat writerName') || pdfOutput

1012

1013 templ <- case templatePath of

1014 _ | not standalone' -> return ""

1015 Nothing -> do

1016 deftemp <- getDefaultTemplate datadir writerName'

1017 case deftemp of

1018 Left e -> throwIO e

1019 Right t -> return t

1020 Just tp -> do

1021 -- strip off extensions

1022 let format = takeWhile (`notElem` "+-") writerName'

1023 let tp' = case takeExtension tp of

1024 "" -> tp <.> format

1025 _ -> tp

1026 E.catch (UTF8.readFile tp')

1027 (\e -> if isDoesNotExistError e

1028 then E.catch

1029 (readDataFileUTF8 datadir

1030 ("templates" </> tp'))

1031 (\e' -> let _ = (e' :: E.SomeException)

1032 in throwIO e')

1033 else throwIO e)

1034

1035 variables' <- case mathMethod of

1036 LaTeXMathML Nothing -> do

1037 s <- readDataFileUTF8 datadir

1038 ("LaTeXMathML.js")

1039 return $ ("mathml-script", s) : variables

1040 MathML Nothing -> do

1041 s <- readDataFileUTF8 datadir

1042 ("MathMLinHTML.js")

1043 return $ ("mathml-script", s) : variables

1044 _ -> return variables

1045

1046 variables'' <- if "dzslides" `isPrefixOf` writerName'

1047 then do

1048 dztempl <- readDataFileUTF8 datadir

1049 ("dzslides" </> "template.html")

1050 let dzcore = unlines $ dropWhile (not . isPrefixOf "<!-- {{{{ dzslides core")

1051 $ lines dztempl

1052 return $ ("dzslides-core", dzcore) : variables'

1053 else return variables'

1054 let sourceURL = case sources of

1055 [] -> Nothing

1056 (x:_) -> case parseURI x of

1057 Just u

1058 | uriScheme u `elem` ["http:","https:"] ->

1059 Just $ show u{ uriPath = "",

1060 uriQuery = "",

1061 uriFragment = "" }

1062 _ -> Nothing

1063

1064 let readerOpts = def{ readerSmart = smart || (texLigatures &&

1065 (laTeXOutput || "context" `isPrefixOf` writerName'))

1066 , readerStandalone = standalone'

1067 , readerParseRaw = parseRaw

1068 , readerColumns = columns

1069 , readerTabStop = tabStop

1070 , readerOldDashes = oldDashes

1071 , readerIndentedCodeClasses = codeBlockClasses

1072 , readerApplyMacros = not laTeXOutput

1073 , readerDefaultImageExtension = defaultImageExtension

1074 }

1075

1076 let writerOptions = def { writerStandalone = standalone',

1077 writerTemplate = templ,

1078 writerVariables = variables'',

1079 writerTabStop = tabStop,

1080 writerTableOfContents = toc,

1081 writerHTMLMathMethod = mathMethod,

1082 writerIncremental = incremental,

1083 writerCiteMethod = citeMethod,

1084 writerIgnoreNotes = False,

1085 writerNumberSections = numberSections,

1086 writerNumberOffset = numberFrom,

1087 writerSectionDivs = sectionDivs,

1088 writerReferenceLinks = referenceLinks,

1089 writerWrapText = wrap,

1090 writerColumns = columns,

1091 writerEmailObfuscation = obfuscationMethod,

1092 writerIdentifierPrefix = idPrefix,

1093 writerSourceURL = sourceURL,

1094 writerUserDataDir = datadir,

1095 writerHtml5 = html5,

1096 writerHtmlQTags = htmlQTags,

1097 writerChapters = chapters,

1098 writerListings = listings,

1099 writerBeamer = False,

1100 writerSlideLevel = slideLevel,

1101 writerHighlight = highlight,

1102 writerHighlightStyle = highlightStyle,

1103 writerSetextHeaders = setextHeaders,

1104 writerTeXLigatures = texLigatures,

1105 writerEpubMetadata = epubMetadata,

1106 writerEpubStylesheet = epubStylesheet,

1107 writerEpubFonts = epubFonts,

1108 writerEpubChapterLevel = epubChapterLevel,

1109 writerTOCDepth = epubTOCDepth,

1110 writerReferenceODT = referenceODT,

1111 writerReferenceDocx = referenceDocx

1112 }

1113

1114 when (not (isTextFormat writerName') && outputFile == "-") $

1115 err 5 $ "Cannot write " ++ writerName' ++ " output to stdout.\n" ++

1116 "Specify an output file using the -o option."

1117

1118 let readSources [] = mapM readSource ["-"]

1119 readSources srcs = mapM readSource srcs

1120 readSource "-" = UTF8.getContents

1121 readSource src = case parseURI src of

1122 Just u | uriScheme u `elem` ["http:","https:"] ->

1123 readURI u

1124 _ -> UTF8.readFile src

1125 readURI uri = simpleHTTP (mkRequest GET uri) >>= getResponseBody >>=

1126 return . UTF8.toStringLazy -- treat all as UTF8

1127

1128 let convertTabs = tabFilter (if preserveTabs then 0 else tabStop)

1129

1130 let handleIncludes' = if readerName' == "latex" || readerName' == "latex+lhs"

1131 then handleIncludes

1132 else return

1133

1134 doc <- readSources sources >>=

1135 handleIncludes' . convertTabs . intercalate "\n" >>=

1136 reader readerOpts

1137

1138

1139 let doc0 = M.foldWithKey setMeta doc metadata

1140 let doc1 = foldr ($) doc0 transforms

1141 doc2 <- foldrM ($) doc1 $ map ($ [writerName']) plugins

1142

1143 let writeBinary :: B.ByteString -> IO ()

1144 writeBinary = B.writeFile (UTF8.encodePath outputFile)

1145

1146 let writerFn :: FilePath -> String -> IO ()

1147 writerFn "-" = UTF8.putStr

1148 writerFn f = UTF8.writeFile f

1149

1150 case getWriter writerName' of

1151 Left e -> err 9 e

1152 Right (IOStringWriter f) -> f writerOptions doc2 >>= writerFn outputFile

1153 Right (IOByteStringWriter f) -> f writerOptions doc2 >>= writeBinary

1154 Right (PureStringWriter f)

1155 | pdfOutput -> do

1156 res <- makePDF latexEngine f writerOptions doc2

1157 case res of

1158 Right pdf -> writeBinary pdf

1159 Left err' -> err 43 $ UTF8.toStringLazy err'

1160 | otherwise -> selfcontain (f writerOptions doc2 ++

1161 ['\n' | not standalone'])

1162 >>= writerFn outputFile . handleEntities

1163 where htmlFormat = writerName' `elem`

1164 ["html","html+lhs","html5","html5+lhs",

1165 "s5","slidy","slideous","dzslides","revealjs"]

1166 selfcontain = if selfContained && htmlFormat

1167 then makeSelfContained datadir

1168 else return

1169 handleEntities = if htmlFormat && ascii

1170 then toEntities

1171 else id