Review Request 130171: Fix page-jumping-back-issue commonly encountered when in presentation mode

[Albert Astals Cid]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/130171/#review103404
-----------------------------------------------------------



Sorry but no, you can't fix presentation mode by removing code out of pageview that you claim you don't know what it does. Debug more and come up with a better patch.

- Albert Astals Cid


On jul. 3, 2017, 8:54 p.m., Alexander Schlarb wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/130171/
> -----------------------------------------------------------
> 
> (Updated jul. 3, 2017, 8:54 p.m.)
> 
> 
> Review request for Okular.
> 
> 
> Repository: okular
> 
> 
> Description
> -------
> 
> This fixes an issue I've commonly encountered while quickly working myself through large PDFs with many pages (such as [this](https://storage.ninetailed.ninja/index.php/s/FS58uge8yKCvjlB)).
> 
> Steps to reproduce the issue:
> 
>  1. Open Okular
>  2. For easily reproducing the issue change *Settings* ? *Performance* ? *Memory Usage* to *Greedy* (the issue can be reproduced without it, but it's harder an alot more random)
>  3. Open a PDF with many pages
>  4. Immidiately go to presentation mode (*Ctrl*+*Shift*+*P*)
>  5. Click on the screen or press the left-arrow keyboard button
> 
> Expected behaviour:
> 
>  * The next slide is shown
> 
> Actual behaviour:
> 
>  * The next slide *is* indeed shown for a split second before the one is sent back to the first slide (this keeps being the case until all slides have been preloaded, then it stops)
>  * Interestingly enough (and I frankly have no idea why) everything works normally if one quickly double-clicks to skip forward by two slides, slides can be viewed normally after that as long as you don't go back to the second slide again…?
> 
> 
> This patch fixes the problematic behaviour by removing code that looks its a bogus fixi from something that shouldn't be a problem… maybe.
> Unfortunately the one attached comment isn't helpful at all in describing what this check was originally supposed to do – just like the commit message for that change (obtained through `git blame`):
> 
>     commit d2d2fa3b2a4fa8f4b554b87ec13ee5a15b64f35d
>     Author: Mailson Menezes <mailson at gmail.com>
>     Date:   Sun Nov 11 23:09:12 2012 -0300
>     
>         Refactoring PageView
> 
> I therefor can only hope that this solution doesn't break anything. Guidance of somebody with in-depth knowledge of the code-base would be apprechiated on this. :-)
> 
> 
> Diffs
> -----
> 
>   ui/pageview.cpp acacfb90 
> 
> Diff: https://git.reviewboard.kde.org/r/130171/diff/
> 
> 
> Testing
> -------
> 
> Loaded a few different PDFs that had previously exibited the described behaviour and verfied that they are fixed now. Also scrolled through them in non-presentation mode, to verify that nothing else appears to be broken now.
> 
> 
> Thanks,
> 
> Alexander Schlarb
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/okular-devel/attachments/20170703/6a52b50f/attachment.html>