#openscad on 2022-04-20 — irc logs at libera.irclog.whitequark.org

00:00 <AndrewStone[m]> WRT offloading computation from the gui thread, it looks like preventing the user from doing anything in the GUI is protecting some objects from use-after-delete. For example, I can now rotate the view of the old model while the new one is compiling. This is really natural because the user doesn't have to wait around; they can still be working. But... crash!

00:00 <AndrewStone[m]> I'll keep at it

00:40 TheCoffeMaker has quit [Ping timeout: 250 seconds]

00:42 TheCoffeMaker has joined #openscad

00:54 TheCoffeMaker has quit [Ping timeout: 246 seconds]

00:56 TheCoffeMaker has joined #openscad

01:15 ur5us has quit [Ping timeout: 250 seconds]

01:36 KimK has quit [Ping timeout: 276 seconds]

01:37 nedko has quit [Ping timeout: 240 seconds]

01:37 Ekho has quit [Quit: CORE ERROR, SYSTEM HALTED.]

01:40 <AndrewStone[m]> seems like Location objects are being deleted out of under the language tree

01:41 sinned69152 has joined #openscad

01:42 rue_mohr has quit [Ping timeout: 272 seconds]

01:42 wed has quit [Ping timeout: 272 seconds]

01:42 lostapathy has quit [Ping timeout: 272 seconds]

01:42 hvitol has quit [Ping timeout: 272 seconds]

01:42 ubitux has quit [Ping timeout: 272 seconds]

01:42 sinned69152 is now known as sinned6915

01:42 sinned6915 has quit [Ping timeout: 272 seconds]

01:42 markasoftware has quit [Ping timeout: 272 seconds]

01:43 ubitux has joined #openscad

01:43 nedko has joined #openscad

01:43 hvitol has joined #openscad

01:43 markasoftware has joined #openscad

01:43 lostapathy has joined #openscad

01:44 J224267 has joined #openscad

01:48 J2242 has quit [Ping timeout: 250 seconds]

01:48 Ekho has joined #openscad

01:51 rue_mohr has joined #openscad

01:57 LordOfBikes has quit [Ping timeout: 240 seconds]

01:57 wed has joined #openscad

02:03 nedko has quit [Ping timeout: 240 seconds]

02:07 nedko has joined #openscad

02:07 dalias has quit [Ping timeout: 272 seconds]

02:08 StudentWeis has joined #openscad

02:10 StudentWeis has quit [Remote host closed the connection]

02:10 StudentWeis has joined #openscad

02:10 LordOfBikes has joined #openscad

02:12 StudentWeis has quit [Remote host closed the connection]

02:12 StudentWeis has joined #openscad

02:14 StudentWeis has quit [Remote host closed the connection]

02:14 StudentWeis has joined #openscad

02:16 StudentWeis has quit [Client Quit]

02:17 nedko has quit [Ping timeout: 240 seconds]

02:19 nedko has joined #openscad

02:22 dalias has joined #openscad

02:24 ur5us has joined #openscad

02:28 nedko has quit [Ping timeout: 240 seconds]

02:33 nedko has joined #openscad

05:06 aiyion has quit [Ping timeout: 240 seconds]

05:09 aiyion has joined #openscad

05:13 ur5us has quit [Ping timeout: 240 seconds]

05:20 aiyion has quit [Ping timeout: 240 seconds]

05:22 aiyion has joined #openscad

05:38 KimK has joined #openscad

05:58 nanoflite has joined #openscad

06:02 nanoflite has quit [Client Quit]

06:05 nedko has quit [Ping timeout: 240 seconds]

06:20 nedko has joined #openscad

06:33 nanoflite has joined #openscad

06:39 nanoflite has quit [Read error: Connection reset by peer]

06:39 nanoflite has joined #openscad

06:39 la1yv_b has quit [Read error: Connection reset by peer]

06:40 la1yv_b has joined #openscad

07:04 nanoflite has quit [Ping timeout: 246 seconds]

07:12 J224267 is now known as J22

07:14 <J22> I wonder if there is still someone working on "roof()" - i have to fiddle around every time to find values that doesn't cause errors.

07:15 <J22> (as i am using offsets ± ∓ to round 2D shapes)

07:26 <J22> seems to fail on higher fn numbers - or more precise if points getting close to each other

07:49 lastrodamo has joined #openscad

07:56 <teepee> J22: I think there's a ticket for that issue already but I'm not aware of anyone actively working on that

07:57 <J22> wasn't it a russian developing that nice feature?

08:32 <teepee> yeah, possible I think

09:03 <crazy_imp> teepee: likely (regarding the flag), but no idea why

09:04 <teepee> my spacemouse is sitting here, I hope I'll find some quiet time at some point :)

09:05 * teepee needs more long easter weekends :)

09:26 ur5us has joined #openscad

10:45 ur5us has quit [Ping timeout: 240 seconds]

11:09 wybpip[m] has joined #openscad

11:09 wybpip[m] has left #openscad [#openscad]

11:24 GNUmoon has quit [Remote host closed the connection]

11:24 little_blossom has joined #openscad

11:46 <peepsalot> i made a script that can recreate git mv's on branches, to match with master. it avoids the "CONFLICT (modify/delete)" issues, but replaces them with "CONFLICT (add/add)"

11:48 <peepsalot> so, I can at use the mergetool, but 3-way merge becomes 2-way (common ancestor is empty).

11:48 <peepsalot> s/at/at least/

11:52 <peepsalot> here's the script https://bpa.st/46JQ

11:52 <peepsalot> I didn't feel like manually recordkeeping the entire history of file moves in a script, so it searches for renames with git itself, and can follow multiple renames

11:55 <teepee> nice. maybe we could stash that into the scripts folder, it's a bit of a mess with some obsolete scripts, but it sounds like we may need that for some time

11:57 <peepsalot> more or less what git merge *should* be doing in the first place, but its not integrated enough to really solve the problem. I'm not sure if there's way to call merge at a lower level, where I can just tell it the 3 commit+filename for each file conflict/merge

12:00 <peepsalot> also, it only seems to help with merges, and not rebases

12:07 <peepsalot> its a start i guess, but I'm still not happy about the 2way merge issue

12:23 <peepsalot> its just immeasurably frustrating to me, because git clearly has the data and ability to properly track renames, but refuses to do the obvious, useful thing

12:50 <teepee> yeah, even when saying it's not the default, having this as option would make a lot of sense

12:53 GNUmoon has joined #openscad

13:14 TheAssassin has quit [Ping timeout: 240 seconds]

13:17 TheAssassin has joined #openscad

13:27 knielsen has quit [Ping timeout: 240 seconds]

13:39 myosotis has joined #openscad

14:18 knielsen has joined #openscad

14:44 salvadorrueda has joined #openscad

14:45 salvadorrueda has quit [Client Quit]

15:06 StudentWeis has joined #openscad

15:08 StudentWeis has quit [Remote host closed the connection]

15:08 StudentWeis has joined #openscad

15:10 StudentWeis has quit [Remote host closed the connection]

15:11 StudentWeis has joined #openscad

15:13 StudentWeis has quit [Remote host closed the connection]

15:13 StudentWeis has joined #openscad

15:14 <J22> doubble rightclick bring up the menue but also changes view to show all hidden geometry and don't revert back to normal view

15:19 StudentWeis has quit [Remote host closed the connection]

15:19 StudentWeis has joined #openscad

15:20 <J22> and render now looks wrong https://pasteboard.co/vzcAeJYWClg8.png

15:22 <J22> the export seems to be ok though .. but i have to close SCAD (or i don't know how to reset this view)

15:27 StudentWeis has quit [Remote host closed the connection]

15:28 StudentWeis has joined #openscad

15:30 StudentWeis has quit [Remote host closed the connection]

15:30 StudentWeis has joined #openscad

15:30 fling has joined #openscad

15:32 StudentWeis has quit [Client Quit]

15:35 fling has quit [Remote host closed the connection]

15:35 fling has joined #openscad

15:51 <teepee> I hereby declare double right click a forbidden action ;-)

15:52 <InPhase> Double right-click is that thing I do when I'm tired, leaning over slightly, and miss the left-click section of my touchpad.

15:52 <InPhase> I can't think of a single interface I've used that had double right-click as an interface feature. (At least not that I noticed.)

15:55 <teepee> ok, what OS, what OpenSCAD version is that? does not seem to happen with the nightly build on Debian/Testing

15:55 <teepee> could depend on the GPU driver though

16:01 <InPhase> peepsalot: So next you need to prepare a PR for git?

16:02 <peepsalot> its written in C though... icky

16:02 <InPhase> Right...

16:05 <Scopeuk> C isn't icky its the cleanest assembler dialect ever! :P

16:17 fling_ has joined #openscad

16:17 fling has quit [Ping timeout: 240 seconds]

16:24 fling_ is now known as fling

16:35 fling has quit [Ping timeout: 240 seconds]

16:36 fling has joined #openscad

17:17 arebil has quit [Quit: My keyboard has gone to sleep. ZZZzzz…]

17:33 Guest88 has joined #openscad

17:33 <J22> teepee win and 22.03.21 ( so ill update )

17:34 Guest88 has quit [Client Quit]

17:35 <J22> actually i was using my second mouse from far away and had to figure out what caused this - as i didn't intended to double rightclick

17:37 arebil has joined #openscad

17:38 fling has quit [Ping timeout: 240 seconds]

17:39 <J22> and you have to double right click in the empty space outside geometry .. and then zoom out

17:39 <J22> (or change the view )

17:40 fling has joined #openscad

17:42 <J22> a cube() is enough to see the issue

17:44 <J22> seems it flipped normals at least the cube is pink with F12

17:55 arebil has quit [Quit: My keyboard has gone to sleep. ZZZzzz…]

18:17 linext has joined #openscad

18:43 fling has quit [Ping timeout: 240 seconds]

18:44 fling has joined #openscad

18:47 osfesa has joined #openscad

19:01 <teepee> osfesa: looks good :)

19:01 <teepee> seems there's another case for image save too

19:28 fling has quit [Ping timeout: 240 seconds]

19:34 <osfesa> I'll have a look

19:55 ur5us has joined #openscad

19:58 <teepee> osfesa: cool, I found it searching for other places using fileExportedMessage()

19:58 <teepee> the qglview->save() already returns a bool but it's ignored

19:59 <teepee> I have not checked yet if it's actually false == failed to save, but that's likely :)

20:00 <osfesa> Yeah, I found it. It does not apply to CSG case. No idea how the << operator behaves in that case

20:00 <osfesa> QImage::save is clear. I guess Qt does not have such bugs nowadays

20:01 <teepee> the csg seems to handle things already as it's testing the stream and errors if it's not open

20:01 <teepee> so I think that's ok

20:02 <teepee> ah, yes, it's QImage::save() in the end

20:04 <osfesa> Can I add the last change into the Pull Request or do I need a new one?

20:04 <osfesa> https://github.com/osfesa/openscad/commit/9d3b80ef7951c0c90fb6cba0fb4468f2838ac2af

20:04 <teepee> it's already added :)

20:05 <teepee> the pull request tracks the branch, so as you committed the new change to that branch, it automatically gets included in the PR

20:09 <osfesa> great, I'll end up liking Github :D

20:12 <teepee> it can be tricky sometimes, but in general they do polish the ussability and integration quite well

20:13 <teepee> still doing monthly backups, just in case :D

20:14 <teepee> also note to self, start making bread earlier so dinner does not end up at midnight ;-)

20:18 osfesa has quit [Ping timeout: 250 seconds]

20:49 Guest72100 has joined #openscad

21:22 lastrodamo has quit [Quit: Leaving]

21:23 <J22> anyone needing a nipple clamp? https://www.printables.com/model/175034

21:24 <gbruno> [github] t-paul pushed 3 modifications (Merge pull request #4231 from osfesa/issue3963 Fix export showing "finished" when file can't be opened.) https://github.com/openscad/openscad/commit/8238b8ccdb2d9894fe9397dd3af652c3918b070e

21:24 <gbruno> [github] t-paul closed issue #3963 (3mf export "finished" when file can't be opened) https://github.com/openscad/openscad/issues/3963

22:00 myosotis has quit [Quit: myosotis]

23:25 Joel has quit [Quit: ZNC 1.7.5+deb4 - https://znc.in]

23:25 mikolajw has quit [Ping timeout: 240 seconds]

23:27 mikolajw has joined #openscad

23:29 Joel has joined #openscad