From 7972b987d4a79b64c7c1689b14d1240a82d50054 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?=E6=9E=97=E6=AA=8E?= <apple@apples-MacBook-Pro.local>
Date: Wed, 1 Mar 2017 13:18:14 +0800
Subject: [PATCH] Read img tag of cover file if cover is (x)html
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

for example
‘<item id="cover" href="Text/cover.xhtml" media-type="application/xhtml+xml”/>’
---
 cps/epub.py | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/cps/epub.py b/cps/epub.py
index f9d46362..26dc839c 100644
--- a/cps/epub.py
+++ b/cps/epub.py
@@ -54,7 +54,17 @@ def get_epub_info(tmp_file_path, original_file_name, original_file_extension):
     else:
         coversection = tree.xpath("/pkg:package/pkg:manifest/pkg:item[@id='cover']/@href", namespaces=ns)
         if len(coversection) > 0:
-            coverfile = extractCover(zip, coversection[0], coverpath, tmp_file_path)
+        filetype = coversection[0].rsplit('.',1)[-1]
+            if filetype == "xhtml" or filetype == "html": #if cover is (x)html format
+                markup = zip.read(os.path.join(coverpath,coversection[0]))
+                markupTree = etree.fromstring(markup)
+                #no matter xhtml or html with no namespace
+                imgsrc = markupTree.xpath( "//*[local-name() = 'img']/@src")
+                #imgsrc maybe startwith "../"" so fullpath join then relpath to cwd
+                filename = os.path.relpath(os.path.join(os.path.dirname(os.path.join(coverpath,coversection[0])),imgsrc[0]))
+                coverfile = extractCover(zip, filename, "", tmp_file_path)        
+            else:
+                coverfile = extractCover(zip, coversection[0], coverpath, tmp_file_path)
         else:
             coverfile = None