PML Input: Various fixes

src/calibre/ebooks/epub/output.py

@@ -148,21 +148,6 @@ class EPUBOutput(OutputFormatPlugin):
                 if not pre.text and len(pre) == 0:
                     pre.tag = 'div'
 
-            # Remove hyperlinks with no content as they cause rendering
-            # artifacts in browser based renderers
-            for a in body.xpath('//a[@href]'):
-                if a.get('id', None) is None and a.get('name', None) is None \
-                        and len(a) == 0 and not a.text:
-                    p = a.getparent()
-                    idx = p.index(a) -1
-                    p.remove(a)
-                    if a.tail:
-                        if idx <= 0:
-                            p.text += a.tail
-                        else:
-                            p[idx].tail += a.tail
-
-
     def convert(self, oeb, output_path, input_plugin, opts, log):
         self.log, self.opts, self.oeb = log, opts, oeb
 

src/calibre/ebooks/oeb/base.py

@@ -904,6 +904,20 @@ class Manifest(object):
                 if key == 'lang' or key.endswith('}lang'):
                     body.attrib.pop(key)
 
+            # Remove hyperlinks with no content as they cause rendering
+            # artifacts in browser based renderers
+            for a in xpath(data, '//h:a[@href]'):
+                if a.get('id', None) is None and a.get('name', None) is None \
+                        and len(a) == 0 and not a.text:
+                    p = a.getparent()
+                    idx = p.index(a) -1
+                    p.remove(a)
+                    if a.tail:
+                        if idx <= 0:
+                            p.text += a.tail
+                        else:
+                            p[idx].tail += a.tail
+
             return data
 
         def _parse_txt(self, data):

src/calibre/ebooks/pml/pmlconverter.py

@@ -43,7 +43,8 @@ PML_HTML_RULES = [
     (re.compile(r'\\-'), lambda match: ''),
     (re.compile(r'\\Fn="(?P<target>.+?)"(?P<text>.*?)\\Fn'), lambda match: '<a href="#footnote-%s">%s</a>' % (match.group('target'), match.group('text')) if match.group('text') else ''),
     (re.compile(r'\\Sd="(?P<target>.+?)"(?P<text>.*?)\\Sd'), lambda match: '<a href="#sidebar-%s">%s</a>' % (match.group('target'), match.group('text')) if match.group('text') else ''),
-    (re.compile(r'\\I'), lambda match: ''),
+    # Just italicize index items as that is how the eReader software renders them.
+    (re.compile(r'\\I(?P<text>.*?)\\I', re.DOTALL), lambda match: '<i>%s</i>' % match.group('text') if match.group('text') else ''),
     
     # Sidebar and Footnotes
     (re.compile(r'<sidebar\s+id="(?P<target>.+?)">\s*(?P<text>.*?)\s*</sidebar>', re.DOTALL), lambda match: '<div id="sidebar-%s">%s</div>' % (match.group('target'), match.group('text')) if match.group('text') else ''),
@@ -55,7 +56,7 @@ PML_HTML_RULES = [
     # Remove empty <p>'s.
     (re.compile('<p>[ ]*</p>'), lambda match: ''),
     # Ensure empty lines carry over.
-    (re.compile('^$', re.MULTILINE), lambda match: '<br />'),
+    (re.compile('(\r\n|\n|\r){3}'), lambda match: '<br />'),
     
     # Remove unmatched plm codes.
     (re.compile(r'(?<=[^\\])\\[pxcriouvtblBk]'), lambda match: ''),