Refactor 2

[popojan]

Refactor seems viable.

• some bug fixes thanks to @lj739's feedback
• undoing moves works again
• komi and engine selection using drop down menus
• komi independent of handicap, possible values listed in data/gui/goban.rml
• fixed secondary engine confusion when switching it during game
• end of game by resignation handled correctly

[lj739]

Please renew the binary file.

------------------ 原始邮件 ------------------ 发件人: "Jan Popelka"<notifications@github.com>; 发送时间: 2020年5月16日(星期六) 晚上8:10 收件人: "popojan/goban"<goban@noreply.github.com>; 抄送: "费而隐"<909417887@qq.com>;"Mention"<mention@noreply.github.com>; 主题: [popojan/goban] Refactor 2 (#20)

Refactor seems viable.

some bug fixes thanks to @lj739's feedback

undoing moves works again

komi and engine selection using drop down menus

komi independent of handicap, possible values listed in data/gui/goban.rml

fixed secondary engine confusion when switching it during game

end of game by resignation handled correctly

You can view, comment on, or merge this pull request online at:

  https://github.com/popojan/goban/pull/20

Commit Summary

configurable komi wip

refactor-2 leftovers

17 drop down komi selection WIP

17 change engines using drop down list

17 fix drop down layout

reenable move sound, fix stuck territory markers

strange behaviour on small screens

18 take back move fixed

15 quick ugly fix

19 all engines follow the game even when inactive

clean some code commented out

16 show influence after resignation

File Changes

 M     CMakeLists.txt     (3)   

 M     data/gui/goban.rcss     (85)   

 M     data/gui/goban.rml     (42)   

 M     src/Board.cpp     (16)   

 M     src/Board.h     (10)   

 M     src/BoardObserver.h     (1)   

 M     src/ElementGame.cpp     (93)   

 M     src/ElementGame.h     (1)   

 M     src/EventHandlerNewGame.cpp     (36)   

 M     src/EventHandlerNewGame.h     (4)   

 M     src/GameObserver.h     (5)   

 M     src/GameState.h     (11)   

 M     src/GameThread.cpp     (184)   

 M     src/GameThread.h     (28)   

 M     src/GobanControl.cpp     (52)   

 M     src/GobanControl.h     (5)   

 M     src/GobanModel.cpp     (68)   

 M     src/GobanModel.h     (25)   

 M     src/GobanView.cpp     (45)   

 M     src/GobanView.h     (4)   

 M     src/opengl.h     (2)   

 M     src/player.cpp     (4)   

 M     src/player.h     (4)   

 M     src/shell/include/ShellOpenGL.h     (2)   

Patch Links:

https://github.com/popojan/goban/pull/20.patch

https://github.com/popojan/goban/pull/20.diff

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[popojan]

Here: hraj.si/goban/goban_2020-05-16_refactor-2.zip. Only goban.exe has changed, though. Thank you for testing!

[popojan]

@lj739 Please tell me if you run into any bugs, in yesterday's build.

[lj739]

I'd given goban to some friends to test, they all acknowledged it is a great interface of go and would have wonderful future. I also let my 7 years old child use it, he like it very much. Since katago support any size of board now, I suppose 15X15 and 17x17 would be welcome. Now goban have had the most important functions already, but there are still many aspects need to be improved. On my view, the first is gtp console, the second, open and save sgf files(including of browse backward and forward in current sgf opened), the third, analyzing.    I have tested the new version for two days, territory still not working, but I can precisely get this information from katago's gtp output log.  

[popojan]

Thank you for your kind words. I'll do my best to come up with some features sooner than in a year ;) Yesterday I compiled katago from sources on my linux dev machine. It works fine even with the integrated Intel HD graphics, so there is no blocker to incorporate some of the great features of katago into goban (except days having only 24 hours, indeed). I may want to ask you for details and discuss the proposed features, so please be around if you can. Have a nice time!

By territory not working you mean showing incorrect ownership or completely broken?

[lj739]

Territory  completely broken, can not show at the end of game, but it doesn't matter. For go players, computing territory is a basic skill,  I suppose  they will not regart it as a  indispensable function, and all they need are katago's gtp console information about winrate and leading stones. Other go interface like sabaki, lizgoban can show  territory at the end of game, but in the middle of game, showing territory is not a reliable thing,  only katago's info can be   relyed on. It is a pleasure to help improving gaban, if you have any question, do not hesitate to contack me. 

------------------ 原始邮件 ------------------ 发件人: "Jan Popelka"<notifications@github.com>; 发送时间: 2020年5月18日(星期一) 晚上7:44 收件人: "popojan/goban"<goban@noreply.github.com>; 抄送: "费而隐"<909417887@qq.com>;"Mention"<mention@noreply.github.com>; 主题: Re: [popojan/goban] Refactor 2 (#20)

Thank you for your kind words. I'll do my best to come up with some features sooner than in a year ;) Yesterday I compiled katago from sources on my linux dev machine. It works fine even with the integrated Intel HD graphics, so there is no blocker to incorporate some of the great features of katago into goban (except days having only 24 hours, indeed). I may want to ask you for details and discuss the proposed features, so please be around if you can. Have a nice time!

By territory not working you mean showing incorrect ownership or completely broken?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.

[popojan]

Thanks for a quick reply, very strange, if you do not touch Territory button, it should show up automatically at the end of the game and disappear when you start a new game. Anytime, you can toggle territory using the button. Midgame you have to wait for a bot move to complete. When game ends by resignation, what goban display is only the dubious mid-game ownership. Let's debug it when you have time, you could submit a new issue, please.

[lj739]

I've  found the reason,  katago engine(1.42 cuda) had caused territory failing to show. On the initial time(no other engines enabled except official engine), territory shown properly. I've got no time to test more engines. Usually I would delete pachi engine, because goban will report a error at launch time if the number of engines exceeds five.

[popojan]

Thanks for narrowing down the problem. I will check it later on my windows box with nvidia card. Now I am reworking and decluttering GUI to make room for GTP console and other new elements.

[popojan]

Already merged as part of gui refactor.