[Poll] Sketcher: Task panel rework.

[openBrain]

This effort, and the corresponding PR's seems to have gone stale. Are the any pending concerns that are still outstanding? I haven't seen any either in this thread or on github.

According to the documents adopted regarding contribution of PR's:

12. Maintainers SHALL merge valid PRs from other Contributors rapidly.
13. Maintainers MAY, at their discretion merge PRs that have not met all criteria to be considered valid to:
1. end fruitless discussions

Have we reached the 'fruitless discussion phase' regarding this? Additionally, it would seem that if merging such a PR into the master dev branch proves to be dysfunctional or largely unpopular, it could be reverted afterwards. Yes? No?

Some remarks aren't took into account.
Commits aren't squashed.
It needs another review after everything in cleaned up.

Contribution process is a whole (and after all maybe only guidelines) that you have to take as such. You can't just excerpt 2 lines and present them as valid by themselves.
To be clear, I have discussed with @abdullah and we decided that we will be pretty finicky on the code that goes to Sketcher given its core importance to FC.

[obelisk79]

Contribution process is a whole (and after all maybe only guidelines) that you have to take as such. You can't just excerpt 2 lines and present them as valid by themselves.
To be clear, I have discussed with @abdullah and we decided that we will be pretty finicky on the code that goes to Sketcher given its core importance to FC.

Thank you for the response. I am not trying to be difficult, just curious because it appeared that the discussion on the PR which I linked went silent for seemingly no reason in November. If the discussion went private, perhaps a status comment on the PR for curious onlookers would help?

[openBrain]

Thank you for the response. I am not trying to be difficult, but because it appeared that the discussion on the PR that I linked went silent for seemingly no reason in November. If the discussion went private, perhaps a status comment on the PR for curious onlookers would help?

No private discussion except a clarification between @abdullah and I about where we want to set the threshold on Sketcher PR so all PRs got an equal processing whoever may merge it.
Actually before Christmas we all got worn on Sketcher PRs (contributors and maintainers) so everyone breathed fresh air and hopefully we can now restart processing them gently.

[obelisk79]

Sounds great, I do follow most conversations and try to only comment when I think there might be some value offered.

I've really been impressed with changes overall since the 0.21 cycle started.

[abdullah]

About Sketcher WB PRs, there are currently:
- 3 PRs open relating to B-Spline support (Ajinkya)
- 3 PRs for sketcher ui (Paddle)
- 1 PR for sketcher ui (Uwe)
- 3 PRs about new tools for Sketcher (Paddle)

Over Christmas I was abroad and GitHub blocked my new device, so instead of merging, I did bug fixing and started developing a Notification Area for FreeCAD.

My absolute priority at the moment is B-Splines. Before Christmas it was merged the ability to constraint B-Spline knots and the ability to set tangency constraints on knot points. I want to merge Point on Object for B-Splines. I had it scheduled for this week. However, there were some bugs in corner cases of the previous merged B-Spline functionalities, I decided to have those bugs fixed first.

The Notification Area should see the light during this month, as I want to get rid of most sketcher unnecessarily blocking notifications.

Then there are the Sketcher UI PRs. As @openBrain says, we all got near burn out with the discussions about these PRs. Discussions about functionality, but also about the internal architecture of the Sketcher. Here I do understand that users following the thread have seen screenshots of functionality and that they wonder why if something appears to work is simply forgotten.

Well, first it is not forgotten. However, it is indeed on hold. It is on hold because sometimes it is not fully working as intended, others because many of the design decisions are, from an architecture point of view, not acceptable. There was strong reluctance from the author of the PRs to accept necessary changes. Often when performing necessary changes, new features were added, which lead to new discussions. I accept my mistakes over the process, as I should have been more active, but I lacked the time.

More important than looking in retrospective is to define how we will move on from here.

I want to reuse what is reusable from the discussions and the code that has been contributed. I will take over this PR, after the B-Spline PRs and the Notification Area are merged. Here, I want to keep the functionality, but change the architecture.

Once this PR is in an acceptable form, the other PRs about UI should be made fit.

Then, later on, I would come back to the functionality PRs.

[Hologram]

I was wondering, if I want to make use of the latest features that have already been merged for my project, what would be the best course of action for me:

• Use 0.20.2 (assuming it is more stable, but less 'feature complete'
  Try master (is it reasonable to assume it's relatively stable?)
  Use Realthunder's branch

I don't see any listing of B-spline or constraints on Realthunder's releases page, so I would assume I would have to try master then. How stable can I expect it to be? I do understand that experimental work shown here on the forums won't be included, but it makes me exited to see such things come to FreeCAD in the future! So I would also like to thank all of the developers for their work.

[drmacro]

I was wondering, if I want to make use of the latest features that have already been merged for my project, what would be the best course of action for me:

• Use 0.20.2 (assuming it is more stable, but less 'feature complete'
  Try master (is it reasonable to assume it's relatively stable?)
  Use Realthunder's branch

I don't see any listing of B-spline or constraints on Realthunder's releases page, so I would assume I would have to try master then. How stable can I expect it to be? I do understand that experimental work shown here on the forums won't be included, but it makes me exited to see such things come to FreeCAD in the future! So I would also like to thank all of the developers for their work.

It is most likely not going to be in 0.20.2. I think chris was referring to 0.21.0 (and it is not in today's git, so I don't know what version he was referring to).
(note: I don't bother with Linkstage, so no clue if or when there...)

As for 0.21.0, it's stable for me, been using it for over a year now.

[chrisb]

My last reply in this topic was from november. In another topic I wrote about the new point-on-B-spline constraint, which is in the current 0.21 version 31641.

Use 0.20.2 (assuming it is more stable, but less 'feature complete'
Try master (is it reasonable to assume it's relatively stable?)

In fact is the current0.21 at least as stable as 0.20.2. The latter is called stable, because it doesn't change every day or week.

[drmacro]

My last reply in this topic was from november. In another topic I wrote about the new point-on-B-spline constraint, which is in the current 0.21 version 31641.

...

I see I was expecting an icon or something. I see it is another possibility with point on object.

[abdullah]

Well, in the meantime some PRs have been merged, and the issue with the DoF abbreviation comes back on the table.

Basically the situation is:
1) Uwe proposed to expand the string "DoF" to "Degrees of Freedom" as the tooltip was useless (#dofs)
2) Paddle did quite some job to reduce the width of the taskbar, which would be in vain if "DoF" gets expanded.

To balance both, I have just improved the tooltips:
https://github.com/FreeCAD/FreeCAD/pull/9343

They now look like this:

Screenshot_20230421_185532.png (41.07 KiB) Viewed 731 times

Screenshot_20230421_185438.png (37.45 KiB) Viewed 731 times

I only put two, of course the other messages also have one appropriate tooltip.

I hope this only leaves some unhappy and nobody torn.