Created attachment 118720 [details] Example slide show Steps to reproduce. 1. Open the attachment in Impress 4.3 2. Start slide show 3. Open the attachment in Impress 5.1 4. Start slide show With hardware acceleration black boxes show up around the transitions. With it disabled, white boxes show up. In 4.3 renders the slideshow correctly.
Adding Michael, since he's been very active in this area recently.
I was able to reproduce this with both Linux and Windows 64bit builds. With the builds I had on hand, I was able to narrow the regression window down to a week. First Bad commit- Build ID: 5191fcb11688ef8f4ec312293e57e151762c6952 2015-08-29 22:35:44 (GMT) Last Good commit- Build ID: f2f97ec5bf4f2808544bef5eb583ed4d424acfa9 015-08-22 19:10:46 (GMT)
git log --oneline f2f97ec5bf4f2808544bef5eb583ed4d424acfa9..5191fcb11688ef8f4ec312293e57e151762c6952 -- vcl | nl Has 67 commits in the range. And quite a lot changed here. Then again - I am assuming you're not using OpenGL acceleration - so that narrows it further. Can you confirm you don't have "use OpenGL" or "force OpenGL" enabled ? =) that will help narrow this down a lot. Thanks !
Michael, "use OpenGL" and "force OpenGL" are both disabled/unchecked. Many devs working in this area. Should I take you off the CC list?
So - it could well still be related =) if you can bisect it further down to a single commit - that would be awesome - then we can find someone to authoritatively blame =)
I git bisected this down to: commit ead5bc3cfb07a4e96e367e7904dc674ee5f5ccd6 Author: Daniel Robertson <danlrobertson89@gmail.com> Date: Fri Aug 21 21:51:09 2015 -0400 slideshow: replace for_each with range-based loop Replace ::std::for_each for a more readable range-based for loop in cases in which the function object to be applied by for_each is more readable as the body of a for loop. Daniel, Can you please take a look at this? There's got to be an issue with the pattern or copy/pasting. It's doing a lot more than making the code more readable:)
Nice work; thanks Luke ! =)
Daniel - any movement here ? Thorsten ?
Thanks Luke for bisecting this! > Daniel - any movement here ? Thorsten ? I've spoken to Luke and Kendy over email. Kendy stated he would revert the commit in the master to close the bug quickly. I will "re-write" ead5bc to be more like 391f18. In addition, after reviewing ead5bc, I'm all over the place. I have changes that are actually addressing https://bugs.documentfoundation.org/show_bug.cgi?id=93243 and I have other random changes in there that are unrelated to the commit subject. I'll split ead5bc into several smaller semantically correct :-) patches.
Jan Holesovsky committed a patch related to this issue. It has been pushed to "master": http://cgit.freedesktop.org/libreoffice/core/commit/?id=862bd364390a05ad0b07f9280fb9a0f8b2cc4ed6 tdf#94222: Revert "slideshow: replace for_each with range-based loop" It will be available in 5.1.0. The patch should be included in the daily builds available at http://dev-builds.libreoffice.org/daily/ in the next 24-48 hours. More information about daily builds can be found at: http://wiki.documentfoundation.org/Testing_Daily_Builds Affected users are encouraged to test the fix and report feedback.
Luke, Daniel: Thank you both! The patch is now reverted, please check it really helped.
*** Bug 94864 has been marked as a duplicate of this bug. ***
Verified fixed in Build ID: 7e48533c21a015d19910373d5c7fd9feaa416622 Thanks Jan!
*** Bug 94875 has been marked as a duplicate of this bug. ***
Version set from description. Best regards. JBF