HTML Input: Ignore unparseable URLs instead of crashing on them. Fixes #902372 (HTML convert crashes with " Invalid IPv6 URL" error)

src/calibre/ebooks/html/input.py

@@ -148,7 +148,11 @@ def find_links(self, src):
                 url = match.group(i)
                 if url:
                     break
-            link = self.resolve(url)
+            try:
+                link = self.resolve(url)
+            except ValueError:
+                # Unparseable URL, ignore
+                continue
             if link not in self.links:
                 self.links.append(link)
 

src/calibre/ebooks/mobi/writer2/serializer.py

@@ -178,7 +178,11 @@ def serialize_href(self, href, base=None):
         at the end.
         '''
         hrefs = self.oeb.manifest.hrefs
-        path, frag = urldefrag(urlnormalize(href))
+        try:
+            path, frag = urldefrag(urlnormalize(href))
+        except ValueError:
+            # Unparseable URL
+            return False
         if path and base:
             path = base.abshref(path)
         if path and path not in hrefs:

src/calibre/ebooks/oeb/transforms/split.py

@@ -154,7 +154,11 @@ def fix_links(self):
 
     def rewrite_links(self, url):
         href, frag = urldefrag(url)
-        href = self.current_item.abshref(href)
+        try:
+            href = self.current_item.abshref(href)
+        except ValueError:
+            # Unparseable URL
+            return url
         if href in self.map:
             anchor_map = self.map[href]
             nhref = anchor_map[frag if frag else None]