https://bugs.documentfoundation.org/show_bug.cgi?id=161898
--- Comment #7 from Manu <[email protected]> --- (In reply to V Stuart Foote from comment #5) Thank you a lot for your participation in this investigation!!! > (In reply to Manu from comment #2) > > Video mkv made with OBS, but impossible to attach here. > > Just reuse the "Add an attachment" link found above, the video will be > accepted if it is not larger than 10MB (I think that is the limit), resample > a smaller section of display or reduce resolution of saved capture if needed. The file is 770ko, so the "(File size limit: 30000 KB)" is ok. In fact the problem was the content type. It is asked this: "Otherwise, choose a method for determining the content type. auto-detect select from list: plain text (text/plain) enter manually: " As there was a preselected "plain text", I wanted to select video (missing in the list), then I wrote "video" manually and I received this warning message: "The content type video is invalid. Valid types must be of the form foo/bar where foo is one of application, audio, image, message, model, multipart, text, video and bar must not contain any special characters (such as "=", "?", ...)." So I thought video was previously accepted (like citation in the warning message), but obviously not anymore accepted for any reason. Today, thanks to your help, I just let the auto-detect activated, and it accepted the file. > Also can not reproduce with steps of OP and attachment 195111 [details], nor > with steps of comment 3 and attachment 195132 [details]. Flash of the status > bar, assume that is the optimal height recalculation. But no visual delay > nor is the message readable. > > You said "not reproduce", but also "flash of the status bar". If you see the flash, then you have reproduced the bug. That's why I explained how to disactivate. Because if the flash was normal, then you shouldn't be able to disactivate it. Imagine the sheet is long to recalculate, and each move between cells recalculate the sheet, this is the long delay that lead me to investigate the problem. I can accept ONE recalculation if needed. But is it really necessary each time I move to another cell without modifying anything? At least now, we have a trick to stop the bug. So, if really nothing can be done to solve the problem, maybe the trick for disactivation manually should be added in the help or somewhere in order to help other users when they are in front of the doom loop adapting the height of lines... maybe also when the delay is very long, a warning box should be displayed in order to help the user to disactivate the problem? -- You are receiving this mail because: You are the assignee for the bug.
