From 90374d24c4a59759dff8dde2ddb3e0cdf39d350c Mon Sep 17 00:00:00 2001
From: Kovid Goyal <kovid@kovidgoyal.net>
Date: Sat, 11 May 2013 13:01:07 +0530
Subject: [PATCH] Dump docx files with prettified xml

---
 src/calibre/ebooks/docx/dump.py | 37 +++++++++++++++++++++++++++++++++
 1 file changed, 37 insertions(+)
 create mode 100644 src/calibre/ebooks/docx/dump.py

diff --git a/src/calibre/ebooks/docx/dump.py b/src/calibre/ebooks/docx/dump.py
new file mode 100644
index 0000000000..f6432125c5
--- /dev/null
+++ b/src/calibre/ebooks/docx/dump.py
@@ -0,0 +1,37 @@
+#!/usr/bin/env python
+# vim:fileencoding=utf-8
+from __future__ import (unicode_literals, division, absolute_import,
+                        print_function)
+
+__license__ = 'GPL v3'
+__copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
+
+import sys, os, shutil
+
+from lxml import etree
+
+from calibre import walk
+from calibre.utils.zipfile import ZipFile
+
+def dump(path):
+    dest = os.path.splitext(os.path.basename(path))[0]
+    dest += '_extracted'
+    if os.path.exists(dest):
+        shutil.rmtree(dest)
+    with ZipFile(path) as zf:
+        zf.extractall(dest)
+
+    for f in walk(dest):
+        if f.endswith('.xml'):
+            with open(f, 'r+b') as stream:
+                raw = stream.read()
+                root = etree.fromstring(raw)
+                stream.seek(0)
+                stream.truncate()
+                stream.write(etree.tostring(root, pretty_print=True, encoding='utf-8', xml_declaration=True))
+
+    print (path, 'dumped to', dest)
+
+if __name__ == '__main__':
+    dump(sys.argv[-1])
+