Add rebase_relative_paths extension.

- Add manual entry for (non-default) extension
  `rebase_relative_paths`.
- Add constructor `Ext_rebase_relative_paths` to `Extensions`
  in Text.Pandoc.Extensions [API change]. When enabled, this
  extension rewrites relative image and link paths by prepending
  the (relative) directory of the containing file.
- Make Markdown reader sensitive to the new extension.
- Add tests for #3752.

Closes #3752.

NB. currently the extension applies to markdown and associated
readers but not commonmark/gfm.

MANUAL.txt

@@ -3755,7 +3755,7 @@ definition:
 Note that space between items in a definition list is required.
 (A variant that loosens this requirement, but disallows "lazy"
 hard wrapping, can be activated with `compact_definition_lists`: see
-[Non-pandoc extensions], below.)
+[Non-default extensions], below.)
 
 [^3]:  I have been influenced by the suggestions of [David
   Wheeler](https://justatheory.com/2009/02/modest-markdown-proposal/).
@@ -5051,13 +5051,37 @@ author-in-text style inside notes when using a note style.
 [finding and editing styles]: https://citationstyles.org/authors/
 [CSL locale files]: https://github.com/citation-style-language/locales
 
-## Non-pandoc extensions
+## Non-default extensions
 
 The following Markdown syntax extensions are not enabled by default
 in pandoc, but may be enabled by adding `+EXTENSION` to the format
 name, where `EXTENSION` is the name of the extension.  Thus, for
 example, `markdown+hard_line_breaks` is Markdown with hard line breaks.
 
+#### Extension: `rebase_relative_paths` ####
+
+Rewrite relative paths for Markdown links and images, depending
+on the path of the file containing the link or image link.  For
+each link or image, pandoc will compute the directory of the
+containing file, relative to the working directory, and prepend
+the resulting path to the link or image path.
+
+The use of this extension is best understood by example.
+Suppose you have a a subdirectory for each chapter of a book,
+`chap1`, `chap2`, `chap3`. Each contains a file `text.md` and a
+number of images used in the chapter.  You would like to have
+`![image](spider.jpg)` in `chap1/text.md` refer to
+`chap1/spider.jpg` and `![image](spider.jpg)` in `chap2/text.md`
+refer to `chap2/spider.jpg`.  To do this, use
+
+    pandoc chap*/*.md -f markdown+rebase_relative_paths
+
+Without this extension, you would have to use
+`![image](chap1/spider.jpg)` in `chap1/text.md` and
+`![image](chap2/spider.jpg)` in `chap2/text.md`.  Links with
+relative paths will be rewritten in the same way as images.
+*This option currently only affects Markdown input.*
+
 #### Extension: `attributes` ####
 
 Allows attributes to be attached to any inline or block-level

pandoc.cabal

@@ -214,6 +214,10 @@ extra-source-files:
                  test/command/C.txt
                  test/command/D.txt
                  test/command/01.csv
+                 test/command/chap1/spider.png
+                 test/command/chap2/spider.png
+                 test/command/chap1/text.md
+                 test/command/chap2/text.md
                  test/command/defaults1.yaml
                  test/command/defaults2.yaml
                  test/command/defaults3.yaml

src/Text/Pandoc/Extensions.hs

@@ -136,6 +136,8 @@ data Extension =
     | Ext_raw_html            -- ^ Allow raw HTML
     | Ext_raw_tex             -- ^ Allow raw TeX (other than math)
     | Ext_raw_markdown        -- ^ Parse markdown in ipynb as raw markdown
+    | Ext_rebase_relative_paths -- ^ Rebase relative image and link paths,
+                                -- relative to directory of containing file
     | Ext_shortcut_reference_links -- ^ Shortcut reference links
     | Ext_simple_tables       -- ^ Pandoc-style simple tables
     | Ext_smart               -- ^ "Smart" quotes, apostrophes, ellipses, dashes
@@ -462,6 +464,7 @@ getAllExtensions f = universalExtensions <> getAll f
        , Ext_gutenberg
        , Ext_smart
        , Ext_literate_haskell
+       , Ext_rebase_relative_paths
        ]
   getAll "markdown_strict"   = allMarkdownExtensions
   getAll "markdown_phpextra" = allMarkdownExtensions

src/Text/Pandoc/Readers/Markdown.hs

@@ -29,7 +29,7 @@ import qualified Data.Set as Set
 import Data.Text (Text)
 import qualified Data.Text as T
 import qualified Data.ByteString.Lazy as BL
-import System.FilePath (addExtension, takeExtension)
+import System.FilePath (addExtension, takeExtension, isAbsolute, takeDirectory)
 import Text.HTML.TagSoup hiding (Row)
 import Text.Pandoc.Builder (Blocks, Inlines)
 import qualified Text.Pandoc.Builder as B
@@ -1836,9 +1836,12 @@ regLink :: PandocMonad m
         -> MarkdownParser m (F Inlines)
 regLink constructor lab = try $ do
   (src, tit) <- source
+  rebase <- option False (True <$ guardEnabled Ext_rebase_relative_paths)
+  pos <- getPosition
+  let src' = if rebase then rebasePath pos src else src
   attr <- option nullAttr $
           guardEnabled Ext_link_attributes >> attributes
-  return $ constructor attr src tit <$> lab
+  return $ constructor attr src' tit <$> lab
 
 -- a link like [this][ref] or [this][] or [this]
 referenceLink :: PandocMonad m
@@ -1854,6 +1857,8 @@ referenceLink constructor (lab, raw) = do
                          return (mempty, "")))
       <|>
       try ((guardDisabled Ext_spaced_reference_links <|> spnl) >> reference)
+  rebase <- option False (True <$ guardEnabled Ext_rebase_relative_paths)
+  pos <- getPosition
   when (raw' == "") $ guardEnabled Ext_shortcut_reference_links
   let labIsRef = raw' == "" || raw' == "[]"
   let key = toKey $ if labIsRef then raw else raw'
@@ -1878,7 +1883,9 @@ referenceLink constructor (lab, raw) = do
                    Just ((src, tit), _) -> constructor nullAttr src tit <$> lab
                    Nothing              -> makeFallback
             else makeFallback
-       Just ((src,tit), attr) -> constructor attr src tit <$> lab
+       Just ((src,tit), attr) ->
+          let src' = if rebase then rebasePath pos src else src
+           in constructor attr src' tit <$> lab
 
 dropBrackets :: Text -> Text
 dropBrackets = dropRB . dropLB
@@ -1911,12 +1918,27 @@ autoLink = try $ do
   return $ return $ B.linkWith attr (src <> escapeURI extra) ""
                      (B.str $ orig <> extra)
 
+-- | Rebase a relative path, by adding the (relative) directory
+-- of the containing source position.  Absolute links and URLs
+-- are untouched.
+rebasePath :: SourcePos -> Text -> Text
+rebasePath pos path = do
+  let fp = sourceName pos
+   in if isAbsolute (T.unpack path) || isURI path
+         then path
+         else
+           case takeDirectory fp of
+             ""  -> path
+             "." -> path
+             d   -> T.pack d <> "/" <> path
+
 image :: PandocMonad m => MarkdownParser m (F Inlines)
 image = try $ do
   char '!'
   (lab,raw) <- reference
   defaultExt <- getOption readerDefaultImageExtension
-  let constructor attr' src = case takeExtension (T.unpack src) of
+  let constructor attr' src =
+         case takeExtension (T.unpack src) of
             "" -> B.imageWith attr' (T.pack $ addExtension (T.unpack src)
                                             $ T.unpack defaultExt)
             _  -> B.imageWith attr' src

test/command/3752.md

@@ -0,0 +1,35 @@
+```
+% pandoc command/chap1/text.md command/chap2/text.md -f markdown+rebase_relative_paths --verbose -t docx | pandoc -f docx -t plain
+^D
+[INFO] Loaded command/chap1/spider.png from ./command/chap1/spider.png
+[INFO] Loaded command/chap1/../../lalune.jpg from ./command/chap1/../../lalune.jpg
+[INFO] Loaded command/chap2/spider.png from ./command/chap2/spider.png
+Chapter one
+
+A spider: [spider]
+
+The moon: [moon]
+
+Link to spider picture.
+
+URL left alone: manual.
+
+Absolute path left alone: absolute.
+
+Chapter two
+
+A spider: [spider]
+```
+
+```
+% pandoc command/chap1/text.md command/chap2/text.md -f markdown+rebase_relative_paths -t html
+^D
+<h1 id="chapter-one">Chapter one</h1>
+<p>A spider: <img src="command/chap1/spider.png" alt="spider" /></p>
+<p>The moon: <img src="command/chap1/../../lalune.jpg" alt="moon" /></p>
+<p>Link to <a href="command/chap1/spider.png">spider picture</a>.</p>
+<p>URL left alone: <a href="https://pandoc.org/MANUAL.html">manual</a>.</p>
+<p>Absolute path left alone: <a href="/foo/bar/baz.png">absolute</a>.</p>
+<h1 id="chapter-two">Chapter two</h1>
+<p>A spider: <img src="command/chap2/spider.png" alt="spider" /></p>
+```

test/command/chap1/text.md

@@ -0,0 +1,11 @@
+# Chapter one
+
+A spider: ![spider](spider.png)
+
+The moon: ![moon](../../lalune.jpg)
+
+Link to [spider picture](spider.png).
+
+URL left alone: [manual](https://pandoc.org/MANUAL.html).
+
+Absolute path left alone: [absolute](/foo/bar/baz.png).

test/command/chap2/text.md

@@ -0,0 +1,3 @@
+# Chapter two
+
+A spider: ![spider](spider.png)