On 10/07/2017 06:38 PM, Richard Heck wrote: > On 10/07/2017 06:14 PM, Scott Kostyshak wrote: >> On Sat, Oct 07, 2017 at 10:24:31AM +0000, Jean-Marc Lasgouttes wrote: >>> Le 07/10/2017 à 03:19, Scott Kostyshak a écrit : >>>> On Fri, Oct 06, 2017 at 08:20:45PM +0000, Richard Heck wrote: >>>>> Recipe: Open a new file. Create a math inset. Type "t" followed by >>>>> "\wed" then hit tab. You will see "\wedge" remaining on-screen as a math >>>>> completion, without the wedge, but the cursor is placed just after where >>>>> the wedge out to be (after the backslash, on my screen). In 2.2.x, the >>>>> wedge just appears. Moreover, if you shift focus away from the LyX >>>>> window, then back, the wedge does appear, but now with "wedge" written >>>>> after it. Similarly with other LaTeX commands, such as \cancel. It seems >>>>> there is some kind of missing update. >>>>> >>>>> You can also get this by typing "ts", then backing up between the t and >>>>> s, and typing "\wed" followed by tab. Oddly, though, if you then hit >>>>> space (or anything), and try "\wed" tab again, it works fine. It's only >>>>> the first time that it fails. But if you create a new math inset, you >>>>> can get it to fail again. >>>> I can reproduce on master but not on 2.3.x. >>> Well, I can reproduce on 2.3.x but not on master ! >> At least we have both covered in terms of reproducibility :) >> >> Note that it is 100% reproducible for me on master, and 0% reproducible >> on 2.3.x. I imagine it is the opposite for you. >> >> My master is a few days old, it is f93ec4a1. Perhaps your >> updateCaretGeometry commit somehow fixed this. >> >> I will compile newest master later. > In master, bisect here blames 764a153c: > > commit 764a153c69bb9b46a6e6872ce48e06f5f867cc53 > Author: Jean-Marc Lasgouttes <lasgout...@lyx.org> > Date: Fri Sep 29 10:29:39 2017 +0200 > > Improve the logic of caret repainting > > For some reason the existing code only considered the bottom row that > contained the cursor. There is no need for that, and actually it > caused painting problems in nested insets. > > Tweak the logic of repaint_caret_row_ a bit: there is no need for > repainting when there is currently no caret on screen.
I'm not sure if it is due to a difference in machines, but I'm now seeing this only on master here, not on 2.3.x. Richard
