From cfd8c0214c3f369d0f8c6f033325c343b78c7659 Mon Sep 17 00:00:00 2001
From: Matthew Pickering <matthewtpickering@gmail.com>
Date: Fri, 8 Aug 2014 23:04:03 +0100
Subject: EPUB Reader: Improved robustness of image extraction

We now maintain the invariant that when fetchImages is called,
all images have absolute paths.

This patch fixes several bugs relating to this as there are three places
where images can be introduced.
  (1) During the HTML parse
  (2) As spine elements
  (3) As a cover image

For (1), the paths are corrected by the transformation renameImages
For (2) and (3), we need to append the "root" to the path we parse from the
spine
---
 src/Text/Pandoc/Readers/EPUB.hs | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

(limited to 'src/Text')

diff --git a/src/Text/Pandoc/Readers/EPUB.hs b/src/Text/Pandoc/Readers/EPUB.hs
index 2967bd09a..8073f9ad2 100644
--- a/src/Text/Pandoc/Readers/EPUB.hs
+++ b/src/Text/Pandoc/Readers/EPUB.hs
@@ -52,10 +52,12 @@ runEPUB = either error id . runExcept
 --
 archiveToEPUB :: (MonadError String m) => ReaderOptions -> Archive -> m (Pandoc, MediaBag)
 archiveToEPUB (setEPUBOptions -> os) archive = do
+  -- root is path to folder with manifest file in
   (root, content) <- getManifest archive
   meta  <- parseMeta content
   (cover, items) <- parseManifest content
-  let coverDoc = fromMaybe mempty (imageToPandoc <$> cover)
+  -- No need to collapse here as the image path is from the manifest file
+  let coverDoc = fromMaybe mempty (imageToPandoc . (root </>) <$> cover)
   spine <- parseSpine items content
   let escapedSpine = map (escapeURI . takeFileName . fst) spine
   Pandoc _ bs <-
@@ -68,17 +70,17 @@ archiveToEPUB (setEPUBOptions -> os) archive = do
     parseSpineElem :: MonadError String m => FilePath -> (FilePath, MIME) -> m Pandoc
     parseSpineElem (normalise -> r) (normalise -> path, mime) = do
       when (readerTrace os) (traceM path)
-      doc <- mimeToReader mime r path
+      doc <- mimeToReader mime (r </> path)
       let docSpan = B.doc $ B.para $ B.spanWith (takeFileName path, [], []) mempty
       return $ docSpan <> doc
-    mimeToReader :: MonadError String m => MIME -> FilePath -> FilePath ->  m Pandoc
-    mimeToReader "application/xhtml+xml" r path = do
-      fname <- findEntryByPathE (r </> path) archive
-      return $ fixInternalReferences (r </> path) .
+    mimeToReader :: MonadError String m => MIME -> FilePath ->  m Pandoc
+    mimeToReader "application/xhtml+xml" (normalise -> path) = do
+      fname <- findEntryByPathE path archive
+      return $ fixInternalReferences path .
                 readHtml os .
                   UTF8.toStringLazy $
                     fromEntry fname
-    mimeToReader s _ path
+    mimeToReader s path
       | s `elem` imageMimes = return $ imageToPandoc path
       | otherwise = return $ mempty
 
-- 
cgit v1.2.3