Add new `--chunk-template` option (closes #8581).

* Add `--chunk-template` CLI option, allowing more control over the
  chunk filenames in chunked HTML output.
* Text.Pandoc.App: Add `optChunkTemplate` constructor to Opt [API change].
* Text.Pandoc.Options: add `writerChunkTemplate` contsructor to
  WriterOptions [API change].
* Text.Pandoc.Chunks: add Data, Typeable, Generic instances for
  PathTemplate.

7 files changed, 36 insertions, 2 deletions

diff --git a/MANUAL.txt b/MANUAL.txt
index 751dc2c5b..0d6ce597b 100644
--- a/MANUAL.txt
+++ b/MANUAL.txt
@@ -1261,6 +1261,17 @@ header when requesting a document from a URL:
     level of 2 or 3. For chunked HTML, this option determines
     how much content goes in each "chunk."
 
+`--chunk-template=`*PATHTEMPLATE*
+
+:   Specify a template for the filenames in a `chunkedhtml` document.
+    In the template, `%n` will be replaced by the chunk number (padded
+    with leading 0s to 3 digits), `%s` with the section number of the chunk,
+    `%h` with the heading text (with formatting removed), `%i` with
+    the section identifier. For example, `%section-%s-%i.html` might
+    be resolved to `section-1.1-introduction.html`. The characters
+    `/` and `\` are not allowed in chunk templates and will be
+    ignored. The default is `%s-%i.html`.
+
 `--epub-chapter-level=`*NUMBER*
 
 :   *Deprecated synonym for `--split-level`.*
diff --git a/src/Text/Pandoc/App/CommandLineOptions.hs b/src/Text/Pandoc/App/CommandLineOptions.hs
index 3e8b832d2..b0e54c92d 100644
--- a/src/Text/Pandoc/App/CommandLineOptions.hs
+++ b/src/Text/Pandoc/App/CommandLineOptions.hs
@@ -793,6 +793,13 @@ options =
                  "NUMBER")
                  "" -- "Header level at which to split documents in chunked HTML or EPUB"
 
+    , Option "" ["chunk-template"]
+                 (ReqArg
+                  (\arg opt ->
+                     return opt{ optChunkTemplate = Just (T.pack arg) })
+                 "PATHTEMPLATE")
+                 "" -- "Template for file paths in chunkedhtml"
+
     , Option "" ["epub-chapter-level"]
                  (ReqArg
                   (\arg opt -> do
diff --git a/src/Text/Pandoc/App/Opt.hs b/src/Text/Pandoc/App/Opt.hs
index 762305170..aa24690dd 100644
--- a/src/Text/Pandoc/App/Opt.hs
+++ b/src/Text/Pandoc/App/Opt.hs
@@ -127,6 +127,7 @@ data Opt = Opt
     , optAbbreviations         :: Maybe FilePath -- ^ Path to abbrevs file
     , optReferenceDoc          :: Maybe FilePath -- ^ Path of reference doc
     , optSplitLevel            :: Int     -- ^ Header level at which to split documents in epub and chunkedhtml
+    , optChunkTemplate         :: Maybe Text -- ^ Template to use for chunk filenames
     , optEpubSubdirectory      :: String -- ^ EPUB subdir in OCF container
     , optEpubMetadata          :: Maybe FilePath   -- ^ EPUB metadata
     , optEpubFonts             :: [FilePath] -- ^ EPUB fonts to embed
@@ -209,6 +210,7 @@ instance FromJSON Opt where
        <*> o .:? "reference-doc"
        <*> ((o .:? "split-level") <|> (o .:? "epub-chapter-level"))
              .!= optSplitLevel defaultOpts
+       <*> o .:? "chunk-template"
        <*> o .:? "epub-subdirectory" .!= optEpubSubdirectory defaultOpts
        <*> o .:? "epub-metadata"
        <*> o .:? "epub-fonts" .!= optEpubFonts defaultOpts
@@ -563,6 +565,8 @@ doOpt (k,v) = do
       parseJSON v >>= \x -> return (\o -> o{ optSplitLevel = x })
     "split-level" ->
       parseJSON v >>= \x -> return (\o -> o{ optSplitLevel = x })
+    "chunk-template" ->
+      parseJSON v >>= \x -> return (\o -> o{ optChunkTemplate = Just x })
     "epub-cover-image" ->
       parseJSON v >>= \x ->
              return (\o -> o{ optEpubCoverImage = unpack <$> x })
@@ -740,6 +744,7 @@ defaultOpts = Opt
     , optAbbreviations         = Nothing
     , optReferenceDoc          = Nothing
     , optSplitLevel            = 1
+    , optChunkTemplate         = Nothing
     , optEpubSubdirectory      = "EPUB"
     , optEpubMetadata          = Nothing
     , optEpubFonts             = []
diff --git a/src/Text/Pandoc/App/OutputSettings.hs b/src/Text/Pandoc/App/OutputSettings.hs
index 17df09e9d..525055d72 100644
--- a/src/Text/Pandoc/App/OutputSettings.hs
+++ b/src/Text/Pandoc/App/OutputSettings.hs
@@ -35,6 +35,7 @@ import System.Directory (getCurrentDirectory)
 import System.Exit (exitSuccess)
 import System.FilePath
 import System.IO (stdout)
+import Text.Pandoc.Chunks (PathTemplate(..))
 import Text.Pandoc
 import Text.Pandoc.App.FormatHeuristics (formatFromFilePaths)
 import Text.Pandoc.App.Opt (Opt (..))
@@ -249,6 +250,9 @@ optToOutputSettings scriptingEngine opts = do
         , writerEpubFonts        = optEpubFonts opts
         , writerEpubTitlePage    = optEpubTitlePage opts
         , writerSplitLevel       = optSplitLevel opts
+        , writerChunkTemplate    = maybe (PathTemplate "%s-%i.html")
+                                     PathTemplate
+                                     (optChunkTemplate opts)
         , writerTOCDepth         = optTOCDepth opts
         , writerReferenceDoc     = optReferenceDoc opts
         , writerSyntaxMap        = syntaxMap
diff --git a/src/Text/Pandoc/Chunks.hs b/src/Text/Pandoc/Chunks.hs
index 9ba282999..2447deb72 100644
--- a/src/Text/Pandoc/Chunks.hs
+++ b/src/Text/Pandoc/Chunks.hs
@@ -4,6 +4,7 @@
 {-# LANGUAGE FlexibleInstances #-}
 {-# LANGUAGE FlexibleContexts #-}
 {-# LANGUAGE GeneralizedNewtypeDeriving #-}
+{-# LANGUAGE DeriveDataTypeable #-}
 {-# LANGUAGE DeriveGeneric #-}
 {- |
    Module      : Text.Pandoc.Chunks
@@ -38,6 +39,8 @@ import Data.String (IsString)
 import GHC.Generics (Generic)
 import Text.HTML.TagSoup (Tag (TagOpen), fromAttrib, parseTags)
 import Data.Tree (Tree(..))
+import Data.Data (Data)
+import Data.Typeable (Typeable)
 
 -- | Split 'Pandoc' into 'Chunk's, e.g. for conversion into
 -- a set of HTML pages or EPUB chapters.
@@ -236,6 +239,7 @@ resolvePathTemplate :: PathTemplate
                     -> FilePath
 resolvePathTemplate (PathTemplate templ) chunknum headingText ident secnum =
   T.unpack .
+  T.filter (\c -> c /= '/' && c /= '\\') .
   T.replace "%n" (T.pack $ printf "%03d" chunknum) .
   T.replace "%s" secnum .
   T.replace "%h" headingText .
@@ -253,7 +257,7 @@ resolvePathTemplate (PathTemplate templ) chunknum headingText ident secnum =
 -- @"section-1.2-introduction.html"@.
 newtype PathTemplate =
   PathTemplate { unPathTemplate :: Text }
-  deriving (Show, IsString)
+  deriving (Show, IsString, Data, Typeable, Generic)
 
 -- | A part of a document (typically a chapter or section, or
 -- the part of a section before its subsections).
diff --git a/src/Text/Pandoc/Options.hs b/src/Text/Pandoc/Options.hs
index f671dd279..d609f591d 100644
--- a/src/Text/Pandoc/Options.hs
+++ b/src/Text/Pandoc/Options.hs
@@ -45,6 +45,7 @@ import GHC.Generics (Generic)
 import Skylighting (SyntaxMap, defaultSyntaxMap)
 import Text.DocTemplates (Context(..), Template)
 import Text.Pandoc.Extensions
+import Text.Pandoc.Chunks (PathTemplate)
 import Text.Pandoc.Highlighting (Style, pygments)
 import Text.Pandoc.UTF8 (toStringLazy)
 import Data.Aeson.TH (deriveJSON)
@@ -318,6 +319,7 @@ data WriterOptions = WriterOptions
   , writerEpubFonts         :: [FilePath] -- ^ Paths to fonts to embed
   , writerEpubTitlePage     :: Bool           -- ^ Include title page in epub
   , writerSplitLevel        :: Int        -- ^ Header level at which to split EPUB or chunked HTML into separate files
+  , writerChunkTemplate     :: PathTemplate  -- ^ Template for filenames in chunked HTML
   , writerTOCDepth          :: Int            -- ^ Number of levels to include in TOC
   , writerReferenceDoc      :: Maybe FilePath -- ^ Path to reference document if specified
   , writerReferenceLocation :: ReferenceLocation    -- ^ Location of footnotes and references for writing markdown
@@ -355,6 +357,7 @@ instance Default WriterOptions where
                       , writerEpubFonts        = []
                       , writerEpubTitlePage    = True
                       , writerSplitLevel       = 1
+                      , writerChunkTemplate    = "%s-%i.html"
                       , writerTOCDepth         = 3
                       , writerReferenceDoc     = Nothing
                       , writerReferenceLocation = EndOfDocument
diff --git a/src/Text/Pandoc/Writers/ChunkedHTML.hs b/src/Text/Pandoc/Writers/ChunkedHTML.hs
index b20f160ce..7287f3a9d 100644
--- a/src/Text/Pandoc/Writers/ChunkedHTML.hs
+++ b/src/Text/Pandoc/Writers/ChunkedHTML.hs
@@ -55,7 +55,7 @@ writeChunkedHTML opts (Pandoc meta blocks) = do
   epochtime <- floor <$> getPOSIXTime
   let toMediaEntry (fp, _mt, bs) = toEntry fp epochtime bs
   mediaEntries <- map toMediaEntry . mediaItems <$> getMediaBag
-  let chunkedDoc = splitIntoChunks "%s-%i.html"
+  let chunkedDoc = splitIntoChunks (writerChunkTemplate opts)
                      True
                      (Just 1)
                      (writerSplitLevel opts)