Raise error on unsupported extensions. Closes #4338.

+ An error is now raised if you try to specify (enable or
  disable) an extension that does not affect the given
  format, e.g. `docx+pipe_tables`.

+ The `--list-extensions[=FORMAT]` option now lists only
  extensions that affect the given FORMAT.

+ Text.Pandoc.Error: Add constructors `PandocUnknownReaderError`,
  `PandocUnknownWriterError`, `PandocUnsupportedExtensionError`.
  [API change]

+ Text.Pandoc.Extensions now exports `getAllExtensions`,
  which returns the extensions that affect a given format
  (whether enabled by default or not). [API change]

+ Text.Pandoc.Extensions: change type of `parseFormatSpec`
  from `Either ParseError (String, Extensions -> Extensions)`
  to `Either ParseError (String, [Extension], [Extension])`
  [API change].

+ Text.Pandoc.Readers: change type of `getReader` so it returns
  a value in the PandocMonad instance rather than an Either
  [API change].  Exceptions for unknown formats and unsupported
  extensions are now raised by this function and need not be handled by
  the calling function.

+ Text.Pandoc.Writers: change type of `getWriter` so it returns
  a value in the PandocMonad instance rather than an Either
  [API change].  Exceptions for unknown formats and unsupported
  extensions are now raised by this function and need not be handled by
  the calling function.

1 files changed, 151 insertions, 15 deletions

diff --git a/src/Text/Pandoc/Extensions.hs b/src/Text/Pandoc/Extensions.hs
index 1c787b7d3..d85b26200 100644
--- a/src/Text/Pandoc/Extensions.hs
+++ b/src/Text/Pandoc/Extensions.hs
@@ -26,6 +26,7 @@ module Text.Pandoc.Extensions ( Extension(..)
                               , enableExtension
                               , disableExtension
                               , getDefaultExtensions
+                              , getAllExtensions
                               , pandocExtensions
                               , plainExtensions
                               , strictExtensions
@@ -312,12 +313,12 @@ strictExtensions = extensionsFromList
 
 -- | Default extensions from format-describing string.
 getDefaultExtensions :: String -> Extensions
-getDefaultExtensions "markdown_strict" = strictExtensions
+getDefaultExtensions "markdown_strict"   = strictExtensions
 getDefaultExtensions "markdown_phpextra" = phpMarkdownExtraExtensions
-getDefaultExtensions "markdown_mmd" = multimarkdownExtensions
-getDefaultExtensions "markdown_github" = githubMarkdownExtensions
-getDefaultExtensions "markdown"        = pandocExtensions
-getDefaultExtensions "ipynb"           =
+getDefaultExtensions "markdown_mmd"      = multimarkdownExtensions
+getDefaultExtensions "markdown_github"   = githubMarkdownExtensions
+getDefaultExtensions "markdown"          = pandocExtensions
+getDefaultExtensions "ipynb"             =
   extensionsFromList
     [ Ext_all_symbols_escapable
     , Ext_pipe_tables
@@ -379,16 +380,149 @@ getDefaultExtensions "opml"            = pandocExtensions -- affects notes
 getDefaultExtensions _                 = extensionsFromList
                                           [Ext_auto_identifiers]
 
--- | Parse a format-specifying string into a markup format and a function that
--- takes Extensions and enables and disables extensions as defined in the format
--- spec.
+allMarkdownExtensions :: Extensions
+allMarkdownExtensions =
+  pandocExtensions <>
+    extensionsFromList
+     [ Ext_old_dashes
+     , Ext_angle_brackets_escapable
+     , Ext_lists_without_preceding_blankline
+     , Ext_four_space_rule
+     , Ext_spaced_reference_links
+     , Ext_hard_line_breaks
+     , Ext_ignore_line_breaks
+     , Ext_east_asian_line_breaks
+     , Ext_emoji
+     , Ext_tex_math_single_backslash
+     , Ext_tex_math_double_backslash
+     , Ext_markdown_attribute
+     , Ext_mmd_title_block
+     , Ext_abbreviations
+     , Ext_autolink_bare_uris
+     , Ext_mmd_link_attributes
+     , Ext_mmd_header_identifiers
+     , Ext_compact_definition_lists
+     , Ext_gutenberg
+     , Ext_smart
+     , Ext_literate_haskell
+     ]
+
+
+-- | Get all valid extensions for a format. This is used
+-- mainly in checking format specifications for validity.
+getAllExtensions :: String -> Extensions
+getAllExtensions f = universalExtensions <> getAll f
+ where
+  autoIdExtensions           = extensionsFromList
+    [ Ext_auto_identifiers
+    , Ext_gfm_auto_identifiers
+    , Ext_ascii_identifiers
+    ]
+  universalExtensions        = extensionsFromList
+    [ Ext_east_asian_line_breaks ]
+  getAll "markdown_strict"   = allMarkdownExtensions
+  getAll "markdown_phpextra" = allMarkdownExtensions
+  getAll "markdown_mmd"      = allMarkdownExtensions
+  getAll "markdown_github"   = allMarkdownExtensions
+  getAll "markdown"          = allMarkdownExtensions
+  getAll "ipynb"             = allMarkdownExtensions
+  getAll "docx"            = extensionsFromList
+    [ Ext_empty_paragraphs
+    , Ext_styles
+    ]
+  getAll "opendocument"    = extensionsFromList
+    [ Ext_empty_paragraphs
+    , Ext_native_numbering
+    ]
+  getAll "odt"             = getAll "opendocument" <> autoIdExtensions
+  getAll "muse"            = autoIdExtensions <>
+    extensionsFromList
+    [ Ext_amuse ]
+  getAll "asciidoc"        = autoIdExtensions
+  getAll "plain"           = allMarkdownExtensions
+  getAll "gfm"             = githubMarkdownExtensions <>
+    autoIdExtensions <>
+    extensionsFromList
+    [ Ext_raw_html
+    , Ext_raw_tex            -- only supported in writer (for math)
+    , Ext_hard_line_breaks
+    , Ext_smart
+    ]
+  getAll "commonmark"      = getAll "gfm"
+  getAll "org"             = autoIdExtensions <>
+    extensionsFromList
+    [ Ext_citations
+    , Ext_smart
+    ]
+  getAll "html"            = autoIdExtensions <>
+    extensionsFromList
+    [ Ext_native_divs
+    , Ext_line_blocks
+    , Ext_native_spans
+    , Ext_empty_paragraphs
+    , Ext_raw_html
+    , Ext_raw_tex
+    , Ext_task_lists
+    , Ext_tex_math_dollars
+    , Ext_tex_math_single_backslash
+    , Ext_tex_math_double_backslash
+    , Ext_literate_haskell
+    , Ext_epub_html_exts
+    ]
+  getAll "html4"           = getAll "html"
+  getAll "html5"           = getAll "html"
+  getAll "epub"            = getAll "html"
+  getAll "epub2"           = getAll "epub"
+  getAll "epub3"           = getAll "epub"
+  getAll "latex"           = autoIdExtensions <>
+    extensionsFromList
+    [ Ext_smart
+    , Ext_latex_macros
+    , Ext_raw_tex
+    , Ext_task_lists
+    , Ext_literate_haskell
+    ]
+  getAll "beamer"          = getAll "latex"
+  getAll "context"         = autoIdExtensions <>
+    extensionsFromList
+    [ Ext_smart
+    , Ext_raw_tex
+    , Ext_ntb
+    ]
+  getAll "textile"         = autoIdExtensions <>
+    extensionsFromList
+    [ Ext_old_dashes
+    , Ext_smart
+    , Ext_raw_tex
+    ]
+  getAll "opml"            = allMarkdownExtensions -- affects notes
+  getAll "twiki"           = autoIdExtensions <>
+    extensionsFromList
+    [ Ext_smart ]
+  getAll "vimwiki"         = autoIdExtensions
+  getAll "dokuwiki"        = autoIdExtensions
+  getAll "tikiwiki"        = autoIdExtensions
+  getAll "rst"             = autoIdExtensions <>
+    extensionsFromList
+    [ Ext_smart
+    , Ext_literate_haskell
+    ]
+  getAll "mediawiki"       = autoIdExtensions <>
+    extensionsFromList
+    [ Ext_smart ]
+  getAll _                 = mempty
+
+
+-- | Parse a format-specifying string into a markup format,
+-- a set of extensions to enable, and a set of extensions to disable.
 parseFormatSpec :: String
-                -> Either ParseError (String, Extensions -> Extensions)
+                -> Either ParseError (String, [Extension], [Extension])
 parseFormatSpec = parse formatSpec ""
   where formatSpec = do
           name <- formatName
-          extMods <- many extMod
-          return (name, \x -> foldl (flip ($)) x extMods)
+          (extsToEnable, extsToDisable) <- foldl (flip ($)) ([],[]) <$>
+                                             many extMod
+          return (name, reverse extsToEnable, reverse extsToDisable)
         formatName = many1 $ noneOf "-+"
         extMod = do
           polarity <- oneOf "-+"
@@ -397,10 +531,12 @@ parseFormatSpec = parse formatSpec ""
                        Just n  -> return n
                        Nothing
                          | name == "lhs" -> return Ext_literate_haskell
-                         | otherwise -> Prelude.fail $ "Unknown extension: " ++ name
-          return $ case polarity of
-                        '-' -> disableExtension ext
-                        _   -> enableExtension ext
+                         | otherwise -> Prelude.fail $
+                                          "Unknown extension: " ++ name
+          return $ \(extsToEnable, extsToDisable) ->
+                    case polarity of
+                        '+' -> (ext : extsToEnable, extsToDisable)
+                        _   -> (extsToEnable, ext : extsToDisable)
 
 #ifdef DERIVE_JSON_VIA_TH
 $(deriveJSON defaultOptions ''Extension)