Typst reader: support auto_identifiers extension.

Headings without an explicit label can now be assigned automatic
identifiers based on the heading text, making them linkable in a
generated table of contents.  The extension is available for the
typst reader but is off by default; enable it with
`-f typst+auto_identifiers`.  The related `gfm_auto_identifiers` and
`ascii_identifiers` extensions are also made available.

Closes #11041.

Text.Pandoc.Readers.Typst.Parsing: PState gains sOptions,
sIdentifiers, and sLogMessages fields, and now has
HasReaderOptions, HasIdentifierList, and HasLogMessages
instances, allowing reuse of the shared registerHeader. (Not an
API change.)

Co-Authored-By: Claude <noreply@anthropic.com>

MANUAL.txt

@@ -3673,7 +3673,7 @@ automatically assigned a unique identifier based on the heading text.
 This extension can be enabled/disabled for the following formats:
 
 input formats
-:  `markdown`, `latex`, `rst`, `mediawiki`, `textile`, `man`
+:  `markdown`, `latex`, `rst`, `mediawiki`, `textile`, `man`, `typst`
 
 output formats
 :  `markdown`, `muse`

src/Text/Pandoc/Extensions.hs

@@ -662,7 +662,8 @@ getAllExtensions f = universalExtensions <> getAll f
   getAll "mediawiki"       = autoIdExtensions <>
     extensionsFromList
     [ Ext_smart ]
-  getAll "typst"           = extensionsFromList [Ext_citations, Ext_smart]
+  getAll "typst"           = autoIdExtensions <>
+    extensionsFromList [Ext_citations, Ext_smart]
   getAll "djot"            = extensionsFromList [Ext_sourcepos]
   getAll "man"             = autoIdExtensions
   getAll _                 = mempty

src/Text/Pandoc/Readers/Typst.hs

@@ -31,6 +31,7 @@ import Typst ( parseTypst, evaluateTypst )
 import Text.Pandoc.Error (PandocError(..))
 import Text.Pandoc.Translations (Term(References), translateTerm)
 import Text.Pandoc.Shared (tshow, blocksToInlines)
+import Text.Pandoc.Parsing (registerHeader, reportLogMessages)
 import Control.Monad.Except (throwError)
 import Control.Monad (MonadPlus (mplus), void, guard, foldM)
 import Control.Monad.Trans (lift)
@@ -79,7 +80,7 @@ readTypst opts inp = do
         Right content -> do
           let content' = fixNesting content
           let labs = findLabels [content']
-          runParserT pPandoc defaultPState{ sLabels = labs }
+          runParserT pPandoc defaultPState{ sLabels = labs, sOptions = opts }
             inputName [content'] >>=
               either (throwError . PandocParseError . T.pack . show) pure
 
@@ -188,6 +189,7 @@ pPandoc = do
         (if null keywords
             then id
             else B.setMeta "keywords" keywords) $ meta
+  reportLogMessages
   pure $ Pandoc meta' (B.toList bs)
 
 pBlocks :: PandocMonad m => P m B.Blocks
@@ -264,8 +266,9 @@ blockHandlers = M.fromList
   ,("heading", \_ mbident fields -> do
       body <- getField "body" fields
       lev <- getField "level" fields <|> pure 1
-      B.headerWith (fromMaybe "" mbident,[],[]) lev
-         <$> pWithContents pInlines body)
+      ils <- pWithContents pInlines body
+      attr <- registerHeader (fromMaybe "" mbident,[],[]) ils
+      pure $ B.headerWith attr lev ils)
   ,("quote", \_ _ fields -> do
       getField "block" fields >>= guard
       body <- getField "body" fields >>= pWithContents pBlocks

src/Text/Pandoc/Readers/Typst/Parsing.hs

@@ -16,10 +16,13 @@ module Text.Pandoc.Readers.Typst.Parsing
 where
 import Control.Monad (MonadPlus)
 import Control.Monad.Reader (lift)
+import Data.Default (def)
 import qualified Data.Foldable as F
 import qualified Data.Map as M
 import Data.Maybe (fromMaybe)
 import Data.Sequence (Seq)
+import Data.Set (Set)
+import qualified Data.Set as Set
 import Data.Text (Text)
 import Text.Parsec
     ( ParsecT, getInput, setInput, tokenPrim )
@@ -27,18 +30,38 @@ import Typst.Types
     ( Identifier, Content(Elt), FromVal(..), Val(VNone) )
 import Text.Pandoc.Class.PandocMonad ( PandocMonad, report )
 import Text.Pandoc.Logging (LogMessage(..))
+import Text.Pandoc.Options (ReaderOptions)
+import Text.Pandoc.Parsing.Capabilities
+    ( HasReaderOptions(..), HasIdentifierList(..), HasLogMessages(..) )
 import Text.Pandoc.Definition
 
 data PState = PState
         { sLabels :: [Text]
-        , sMeta :: Meta }
+        , sMeta :: Meta
+        , sOptions :: ReaderOptions
+        , sIdentifiers :: Set Text
+        , sLogMessages :: [LogMessage] }
         deriving (Show)
 
+instance HasReaderOptions PState where
+  extractReaderOptions = sOptions
+
+instance HasIdentifierList PState where
+  extractIdentifierList = sIdentifiers
+  updateIdentifierList f st = st{ sIdentifiers = f (sIdentifiers st) }
+
+instance HasLogMessages PState where
+  addLogMessage m st = st{ sLogMessages = m : sLogMessages st }
+  getLogMessages = reverse . sLogMessages
+
 defaultPState :: PState
 defaultPState =
   PState
   { sLabels = []
-  , sMeta = mempty }
+  , sMeta = mempty
+  , sOptions = def
+  , sIdentifiers = Set.empty
+  , sLogMessages = [] }
 
 type P m a = ParsecT [Content] PState m a
 -- state tracks a list of labels in the document

test/command/typst-auto-identifiers.md

@@ -0,0 +1,37 @@
+With `auto_identifiers` enabled, headings without an explicit label get
+automatic identifiers, with collisions disambiguated:
+
+```
+% pandoc -f typst+auto_identifiers -t native
+= Introduction
+
+Text.
+
+== A Sub Section
+
+= Introduction
+^D
+[ Header
+    1 ( "introduction" , [] , [] ) [ Str "Introduction" ]
+, Para [ Str "Text." ]
+, Header
+    2
+    ( "a-sub-section" , [] , [] )
+    [ Str "A" , Space , Str "Sub" , Space , Str "Section" ]
+, Header
+    1 ( "introduction-1" , [] , [] ) [ Str "Introduction" ]
+]
+```
+
+The extension is off by default, so headings have no identifiers:
+
+```
+% pandoc -f typst -t native
+= Introduction
+
+Text.
+^D
+[ Header 1 ( "" , [] , [] ) [ Str "Introduction" ]
+, Para [ Str "Text." ]
+]
+```