Ensure that valid XML identifiers are used in...

Docbook, EPUB, FB2, HTML4, S5, Slidy, Slideous,
ICML, ODT, TEI writers.

Thus, if you convert `[anchor]{#1} and [link to](#1)`,
`id_1` will be used instead of `1` for the identifier.

src/Text/Pandoc/Writers/Docbook.hs

@@ -94,7 +94,8 @@ writeDocbook5 opts d =
 
 -- | Convert Pandoc document to string in Docbook format.
 writeDocbook :: PandocMonad m => WriterOptions -> Pandoc -> DB m Text
-writeDocbook opts (Pandoc meta blocks) = do
+writeDocbook opts doc = do
+  let Pandoc meta blocks = ensureValidXmlIdentifiers doc
   let colwidth = if writerWrapText opts == WrapAuto
                     then Just $ writerColumns opts
                     else Nothing

src/Text/Pandoc/Writers/EPUB.hs

@@ -35,7 +35,8 @@ import qualified Data.Text.Lazy as TL
 import System.FilePath (takeExtension, takeFileName, makeRelative)
 import Text.HTML.TagSoup (Tag (TagOpen), fromAttrib, parseTags)
 import Text.Pandoc.Builder (fromList, setMeta)
-import Text.Pandoc.Class.PandocMonad (PandocMonad, report)
+import Text.Pandoc.Writers.Shared (ensureValidXmlIdentifiers)
+import Text.Pandoc.Class (PandocMonad, report)
 import qualified Text.Pandoc.Class.PandocPure as P
 import qualified Text.Pandoc.Class.PandocMonad as P
 import Data.Time
@@ -435,8 +436,9 @@ pandocToEPUB :: PandocMonad m
              -> Pandoc
              -> E m B.ByteString
 pandocToEPUB version opts doc = do
+  let doc' = ensureValidXmlIdentifiers doc
   -- handle pictures
-  Pandoc meta blocks <- walkM (transformInline opts) doc >>=
+  Pandoc meta blocks <- walkM (transformInline opts) doc' >>=
                         walkM transformBlock
   picEntries <- mapMaybe (snd . snd) <$> gets stMediaPaths
 

src/Text/Pandoc/Writers/FB2.hs

@@ -40,7 +40,8 @@ import Text.Pandoc.Logging
 import Text.Pandoc.Options (HTMLMathMethod (..), WriterOptions (..), def)
 import Text.Pandoc.Shared (capitalize, isURI, orderedListMarkers,
                            makeSections, tshow, stringify)
-import Text.Pandoc.Writers.Shared (lookupMetaString, toLegacyTable)
+import Text.Pandoc.Writers.Shared (lookupMetaString, toLegacyTable,
+                                   ensureValidXmlIdentifiers)
 import Data.Generics (everywhere, mkT)
 
 -- | Data to be written at the end of the document:
@@ -76,7 +77,8 @@ pandocToFB2 :: PandocMonad m
             => WriterOptions
             -> Pandoc
             -> FBM m Text
-pandocToFB2 opts (Pandoc meta blocks) = do
+pandocToFB2 opts doc = do
+     let Pandoc meta blocks = ensureValidXmlIdentifiers doc
      modify (\s -> s { writerOptions = opts })
      desc <- description meta
      title <- cMapM toXml . docTitle $ meta

src/Text/Pandoc/Writers/HTML.hs

@@ -149,12 +149,14 @@ writeHtml5 = writeHtml' defaultWriterState{ stHtml5 = True }
 
 -- | Convert Pandoc document to Html 4 string.
 writeHtml4String :: PandocMonad m => WriterOptions -> Pandoc -> m Text
-writeHtml4String = writeHtmlString'
-                      defaultWriterState{ stHtml5 = False }
+writeHtml4String opts = writeHtmlString'
+                         defaultWriterState{ stHtml5 = False } opts .
+                        ensureValidXmlIdentifiers
 
 -- | Convert Pandoc document to Html 4 structure.
 writeHtml4 :: PandocMonad m => WriterOptions -> Pandoc -> m Html
-writeHtml4 = writeHtml' defaultWriterState{ stHtml5 = False }
+writeHtml4 opts = writeHtml' defaultWriterState{ stHtml5 = False } opts .
+                    ensureValidXmlIdentifiers
 
 -- | Convert Pandoc document to Html appropriate for an epub version.
 writeHtmlStringForEPUB :: PandocMonad m
@@ -164,6 +166,8 @@ writeHtmlStringForEPUB version o = writeHtmlString'
                       defaultWriterState{ stHtml5 = version == EPUB3,
                                           stEPUBVersion = Just version }
                       o{ writerWrapText = WrapNone }
+   -- we don't use ensureValidXmlIdentifiers here because we
+   -- do that in the EPUB writer
 
 -- | Convert Pandoc document to Reveal JS HTML slide show.
 writeRevealJs :: PandocMonad m
@@ -173,22 +177,25 @@ writeRevealJs = writeHtmlSlideShow' RevealJsSlides
 -- | Convert Pandoc document to S5 HTML slide show.
 writeS5 :: PandocMonad m
         => WriterOptions -> Pandoc -> m Text
-writeS5 = writeHtmlSlideShow' S5Slides
+writeS5 opts = writeHtmlSlideShow' S5Slides opts .
+               ensureValidXmlIdentifiers
 
 -- | Convert Pandoc document to Slidy HTML slide show.
 writeSlidy :: PandocMonad m
            => WriterOptions -> Pandoc -> m Text
-writeSlidy = writeHtmlSlideShow' SlidySlides
+writeSlidy opts = writeHtmlSlideShow' SlidySlides opts .
+                  ensureValidXmlIdentifiers
 
 -- | Convert Pandoc document to Slideous HTML slide show.
 writeSlideous :: PandocMonad m
               => WriterOptions -> Pandoc -> m Text
-writeSlideous = writeHtmlSlideShow' SlideousSlides
+writeSlideous opts = writeHtmlSlideShow' SlideousSlides opts .
+                     ensureValidXmlIdentifiers
 
 -- | Convert Pandoc document to DZSlides HTML slide show.
 writeDZSlides :: PandocMonad m
               => WriterOptions -> Pandoc -> m Text
-writeDZSlides = writeHtmlSlideShow' DZSlides
+writeDZSlides opts = writeHtmlSlideShow' DZSlides opts
 
 writeHtmlSlideShow' :: PandocMonad m
                     => HTMLSlideVariant -> WriterOptions -> Pandoc -> m Text

src/Text/Pandoc/Writers/ICML.hs

@@ -131,7 +131,8 @@ citeName          = "Cite"
 
 -- | Convert Pandoc document to string in ICML format.
 writeICML :: PandocMonad m => WriterOptions -> Pandoc -> m Text
-writeICML opts (Pandoc meta blocks) = do
+writeICML opts doc = do
+  let Pandoc meta blocks = ensureValidXmlIdentifiers doc
   let colwidth = if writerWrapText opts == WrapAuto
                     then Just $ writerColumns opts
                     else Nothing

src/Text/Pandoc/Writers/ODT.hs

@@ -36,7 +36,8 @@ import Text.Pandoc.Options (WrapOption (..), WriterOptions (..))
 import Text.DocLayout
 import Text.Pandoc.Shared (stringify, pandocVersion, tshow)
 import Text.Pandoc.Writers.Shared (lookupMetaString, lookupMetaBlocks,
-                                   fixDisplayMath, getLang)
+                                   fixDisplayMath, getLang,
+                                   ensureValidXmlIdentifiers)
 import Text.Pandoc.UTF8 (fromStringLazy, fromTextLazy, toTextLazy)
 import Text.Pandoc.Walk
 import Text.Pandoc.Writers.OpenDocument (writeOpenDocument)
@@ -58,8 +59,9 @@ writeODT :: PandocMonad m
 writeODT  opts doc =
   let initState = ODTState{ stEntries = []
                           }
+      doc' = ensureValidXmlIdentifiers doc
   in
-    evalStateT (pandocToODT opts doc) initState
+    evalStateT (pandocToODT opts doc') initState
 
 -- | Produce an ODT file from a Pandoc document.
 pandocToODT :: PandocMonad m

src/Text/Pandoc/Writers/TEI.hs

@@ -28,7 +28,8 @@ import Text.Pandoc.XML
 
 -- | Convert Pandoc document to string in Docbook format.
 writeTEI :: PandocMonad m => WriterOptions -> Pandoc -> m Text
-writeTEI opts (Pandoc meta blocks) = do
+writeTEI opts doc = do
+  let Pandoc meta blocks = ensureValidXmlIdentifiers doc
   let colwidth = if writerWrapText opts == WrapAuto
                     then Just $ writerColumns opts
                     else Nothing