Quote:
Originally Posted by tshering
It is not a question of xml validity. The Kobo software expects that those tags are not nested.
|
So
https://bugs.launchpad.net/calibre-k...r/+bug/1205702 should hopefully cover this when I get it done then. Here's hoping
Quote:
Originally Posted by tshering
Why don't you try what I proposed and judge by yourself. My proposal was meant exactly for the test file. Since I propose adding one closing tag at an early position and deleting one later, the balance is still preserved.
|
Only because it's more involved than that if the Kobo software expects the tags are not nested.
Quote:
Originally Posted by tshering
I thought Davidfor might have reported this, cf. this post. |
That may have been on the previous iteration of the algorithm for adding spans. I could never reproduce it with the current version, but maybe I just haven't tried a sufficiently complex book yet. Again, hoping that
https://bugs.launchpad.net/calibre-k...r/+bug/1205702 will deal with it.
Quote:
Originally Posted by PeterT
One more experiment I tried was to change all the span id's to be unique; this helped in the first test document, I could now jump to chapter 3 successfully, but did not help in the second test document.
(I took the brute force approach of changing kobo. to kobo.# where # was the digit on the html file's name).
|
Now that is interesting. I'd been working under the assumption that IDs only had to be unique within a file. I'll make a change first to force unique IDs across the whole book and see if it makes any difference for other people.