Description: I'm using LibreOffice on Arch Linux. My input method framework is fcitx5. Also fcitx4 has the bug. My input method candidate window offset is double of the expected offset relative to screen left top both horizontally and vertically, which moves outside of the screen when I type at the bottom of the screen. After my observation, the offset is accurately doubled and I happen to use a 2-scale display. So I guess it might be that problem. By the way, I use KDE Plasma. Steps to Reproduce: 1. Use a Linux system and use fcitx framework (version 4 or 5 is OK), and any input methods (I tried built-in one and Sougo Pinyin). 2. Open Writer or Calc (Other components may also have the bug.) 3. Type and make it show a candidate window. Actual Results: The candidate window offset is much bigger than expected at approximately double value. Expected Results: The candidate window should be near the cursor. Reproducible: Always User Profile Reset: No Additional Info: Version: 7.1.1.2 / LibreOffice Community Build ID: 10(Build:2) CPU threads: 8; OS: Linux 5.11; UI render: default; VCL: kf5 Locale: en-US (en_US.UTF-8); UI: en-US 7.1.1-1 Calc: threaded
Created attachment 170736 [details] A screenshot showing the bug.
I cannot reproduce with fctix5. Can you try again with a newer version to see if the issue persists? Version: 7.4.0.3 / LibreOffice Community Build ID: 40(Build:3) CPU threads: 4; OS: Linux 5.15; UI render: default; VCL: kf5 (cairo+xcb) Locale: en-GB (en_US.UTF-8); UI: en-US 7.4.0-3 Calc: threaded
Thank you for your reply! I’m sorry that I no longer use that system and environment. It might be difficult for me to examine the bug again. If you can’t reproduce that in a newer version and nobody else is complaining about it, I think it would be OK to close this bug. Maybe we are lucky to have it fixed magically somewhere.
Marking as worksforme because I cannot reproduce and the original reporter is also not in the position to reproduce. If this issue resurfaces for another user, they can open a new bug.