Allow tweaking of DOCX files using calibre-debug --explode-book

src/calibre/ebooks/docx/dump.py

@@ -13,15 +13,8 @@ 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 += '-dumped'
-    if os.path.exists(dest):
-        shutil.rmtree(dest)
-    with ZipFile(path) as zf:
-        zf.extractall(dest)
-
-    for f in walk(dest):
+def pretty_all_xml_in_dir(path):
+    for f in walk(path):
         if f.endswith('.xml') or f.endswith('.rels'):
             with open(f, 'r+b') as stream:
                 raw = stream.read()
@@ -31,6 +24,15 @@ def dump(path):
                     stream.truncate()
                     stream.write(etree.tostring(root, pretty_print=True, encoding='utf-8', xml_declaration=True))
 
+def dump(path):
+    dest = os.path.splitext(os.path.basename(path))[0]
+    dest += '-dumped'
+    if os.path.exists(dest):
+        shutil.rmtree(dest)
+    with ZipFile(path) as zf:
+        zf.extractall(dest)
+    pretty_all_xml_in_dir(dest)
+
     print (path, 'dumped to', dest)
 
 if __name__ == '__main__':

src/calibre/ebooks/tweak.py

@@ -70,6 +70,15 @@ def zip_rebuilder(tdir, path):
                 zfn = unicodedata.normalize('NFC', os.path.relpath(absfn, tdir).replace(os.sep, '/'))
                 zf.write(absfn, zfn)
 
+def docx_exploder(path, tdir, question=lambda x:True):
+    zipextract(path, tdir)
+    from calibre.ebooks.docx.dump import pretty_all_xml_in_dir
+    pretty_all_xml_in_dir(tdir)
+    for f in walk(tdir):
+        if os.path.basename(f) == 'document.xml':
+            return f
+    raise Error('Invalid book: Could not find document.xml')
+
 def get_tools(fmt):
     fmt = fmt.lower()
 
@@ -78,6 +87,8 @@ def get_tools(fmt):
         ans = mobi_exploder, rebuild
     elif fmt in {'epub', 'htmlz'}:
         ans = zip_exploder, zip_rebuilder
+    elif fmt == 'docx':
+        ans = docx_exploder, zip_rebuilder
     else:
         ans = None, None