Re: A fix for abbreviations with foreign keyboards is in the #1563 branch

Hello Brian, hello Edward, Am Sonntag, 10. Mai 2020 20:53:26 UTC+2 schrieb btheado: > > On Sun, May 10, 2020 at 2:22 PM Viktor Ransmayr > wrote: > >> Since it is the first time that I'm trying to work on a Git branch, it >> might well be that the problem is on my side ;-) >> >> The first (strang

Re: A fix for abbreviations with foreign keyboards is in the #1563 branch

Viktor, On Sun, May 10, 2020 at 2:22 PM Viktor Ransmayr wrote: > Since it is the first time that I'm trying to work on a Git branch, it > might well be that the problem is on my side ;-) > > The first (strange?) thing I've to report is that after I did a 'git pull' > - and - tried to do a 'git c

Re: A fix for abbreviations with foreign keyboards is in the #1563 branch

Hello Edward Am Sonntag, 10. Mai 2020 17:05:08 UTC+2 schrieb Edward K. Ream: > > This fix involves a tweak to Leo's lowest level key-handling code, namely > filter.doAltTweaks. No change to Leo's abbreviation code seems to be > necessary. > > Although the change appears slight, it might have sig

A fix for abbreviations with foreign keyboards is in the #1563 branch

This fix involves a tweak to Leo's lowest level key-handling code, namely filter.doAltTweaks. No change to Leo's abbreviation code seems to be necessary. Although the change appears slight, it might have significant ramifications. Indeed, my first attempt (in devel) failed. I backed out of the

Re: ENB: Qt-prototype is finished

> > Can you also add a hypothesis test which instead of executing the random > modification methods one-at-a-time, they get executed via execute_script? > Maybe there are some bugs to shake out there. > > Brian > Thanks, this is a good suggestion. But right now I am considering another idea.

Re: ENB: Qt-prototype is finished

Edward, On Sun, May 10, 2020 at 8:29 AM Edward K. Ream wrote: > Yes that was the idea. Currently I am working on another idea, but it is >> too soon to share it. I'll write more once I have something to show. >> > > This sounds reasonable. As always, I'll reserve judgment until I see code. > I e

Re: Qt-prototype report and some thoughts about executeScript command

On Sun, May 10, 2020 at 8:59 AM vitalije wrote: > I didn't find that hypothesis was slowing me down during development. When > a bug combination is found, hypothesis checks this combination early. It > means that until you fix the bug running hypothesis is almost exactly as > running a special ha

Re: Qt-prototype report and some thoughts about executeScript command

I didn't find that hypothesis was slowing me down during development. When a bug combination is found, hypothesis checks this combination early. It means that until you fix the bug running hypothesis is almost exactly as running a special hand-written test for just this sequence of commands. If

Re: Push to Leo Repo Failed with error:Protected branch update failed for refs/heads/devel.

During the long saga, someone - was it Matt? - suggested removing the credentials helper from git so that user/pw would be required for each push. I did that. Now that it's all working, I was able to restore the helper and I don't need to provide them each time. On Windows, here's the comman

Newbie Aha re Hypothesis draw

Sheesh. The word "draw" in the draw strategy has nothing to do with drawing a screen :-) "Draw" is used as in statistics, for example, "draw one marble from a bag containing 10 white marbles and 5 red marbles". Surprising

Re: Qt-prototype report and some thoughts about executeScript command

On Saturday, May 9, 2020 at 3:48:45 PM UTC-5, vitalije wrote: > Good catch Brian. Revision 8184a2023e5f9 contains the fix for this issue. Now hoist/dehoist commands are also tested using hypothesis. I'm just beginning to see how cool Hypothesis is. All you had to do is add two command names to

Re: ENB: Qt-prototype is finished

On Sun, May 10, 2020 at 5:00 AM vitalije wrote: > >> Could you explain more on how that would work? The outline commands, for >> example, currently live as methods on c. But with your new code they would >> be available at c.guiapi.xxx. The code which currently calls these outline >> commands (m

Re: Qt-prototype report and some thoughts about executeScript command

21ead02d970 fixes empty looking tree. It is still possible to add more siblings of hoisted node, but I don't mind this behavior. On Sunday, May 10, 2020 at 3:02:01 AM UTC+2, btheado wrote: > > It no longer crashes for me, thanks. > > While it doesn't crash, it is still possible to move a node lef

Re: ENB: Qt-prototype is finished

> > > Could you explain more on how that would work? The outline commands, for > example, currently live as methods on c. But with your new code they would > be available at c.guiapi.xxx. The code which currently calls these outline > commands (minibuffer commands and keystroke callbacks?) wou