PML Input: Clean import. PMLZ Output: Put images in DropBook required image subfolder. Remove \Cn tags written inside of \x and \Xn tags.

src/calibre/ebooks/pml/input.py

@@ -1,4 +1,3 @@
-import os.path
 # -*- coding: utf-8 -*-
 
 __license__   = 'GPL v3'

src/calibre/ebooks/pml/output.py

@@ -50,7 +50,10 @@ class PMLOutput(OutputFormatPlugin):
             with open(os.path.join(tdir, 'index.pml'), 'wb') as out:
                 out.write(pml.encode(opts.pml_output_encoding, 'replace'))
 
-            self.write_images(oeb_book.manifest, pmlmlizer.image_hrefs, tdir, opts)
+            img_path = os.path.join(tdir, 'index_img')
+            if not os.path.exists(img_path):
+                os.makedirs(img_path)
+            self.write_images(oeb_book.manifest, pmlmlizer.image_hrefs, img_path, opts)
 
             log.debug('Compressing output...')
             pmlz = ZipFile(output_path, 'w')

src/calibre/ebooks/pml/pmlml.py

@@ -180,6 +180,9 @@ class PMLMLizer(object):
         links = set(re.findall(r'(?<=\\q="#).+?(?=")', text))
         for unused in anchors.difference(links):
             text = text.replace('\\Q="%s"' % unused, '')
+            
+        # Remove \Cn tags that are within \x and \Xn tags
+        text = re.sub(ur'(?msu)(?P<t>\\(x|X[0-4]))(?P<a>.*?)(?P<c>\\C[0-4]\s*=\s*"[^"]*")(?P<b>.*?)(?P=t)', '\g<t>\g<a>\g<b>\g<t>', text)
 
         # Replace bad characters.
         text = text.replace(u'\xc2', '')