Comment 2 for bug 1658578

So, are you saying that you have coded in a change for loading the html due to security reasons in calibre version 2.77 that was not coded into version 2.72? Version 2.72 is lightning fast when changing pages; 2.77 is not.
Can I send you a link to download the epub I am using (via WeTransfer)?
You can test an abridged version that also exhibits the behavior I am describing, downloaded from Smashwords:

https://www.smashwords.com/books/view/682810

But the unabridged file is ten times larger, and much slower.

-----Original Message-----
From: <email address hidden> [mailto:<email address hidden>] On Behalf Of Kovid Goyal
Sent: Monday, January 23, 2017 2:32 PM
To: <email address hidden>
Subject: [Bug 1658578] Re: calibre bug 1658578

Loading flow happens whenever the internal HTML file being displayed is changed, the new HTML file has to be rendered and laid out. That remains the same in all versions of the viewer. If you are saying that the amount of time taken for each individual loading flow has increased, then it is likely because of a change to how the html is loaded which is necessary for security reasons -- nothing I can do to fix that. Although I have to say that I can see no increase in time for each loading flow (it remains imperceptible) on, for example, War and Peace.

--
You received this bug notification because you are subscribed to the bug report.
https://bugs.launchpad.net/bugs/1658578

Title:
  loading flow delay for every page

Status in calibre:
  New

Bug description:
  PC environment: Windows 7 Ultimate 64-bit with 12Gb RAM in system
  calibre version 2.72 experiences the normal "loading flow" delay once when initially viewing an epub, and subsequent page links load immediately. But calibre version 2.77 experiences this "loading flow" delay with each and every page.
  With a small epub, the delay may be acceptable, but with the epub I am using (33,000 xhtml text files, 255,000 pages of text, over one million hyperlinks; no graphics other than the cover), this delay causes calibre to be non-functional. Perhaps a caching function algorithm was changed?
  I can send the file for your testing, if needed (it is a reference dictionary, 200Mb compressed), and I am willing to assist with debugging in any way you may request.

To manage notifications about this bug go to:
https://bugs.launchpad.net/calibre/+bug/1658578/+subscriptions