equim-chan / mjai-reviewer Goto Github PK
View Code? Open in Web Editor NEW🔍🀄️ Review mahjong game log with mjai-compatible mahjong AI.
Home Page: https://mjai.ekyu.moe
License: Apache License 2.0
🔍🀄️ Review mahjong game log with mjai-compatible mahjong AI.
Home Page: https://mjai.ekyu.moe
License: Apache License 2.0
It would be nice to be able to see at a glance how much "points" a mistake has cost vs. the move recommended by Akochan. It would help to quickly find big blunders vs. minor mistakes.
You can currently do the math manually, but it would be more convenient to have the value somewhere easily accessible.
For example, East game log generated by log extractor contains word "East" on rule.disp
.
However, in akochan-reviewer, only word '東' is checked whether the game is Tonpuu (East) game; if not, the log is examined as Hanchan (South) game.
For workaround, I manually modify log file to add word '東' in rule.disp
. But, it would be better if akochan-review could recognize word 'East' as Tonpuu (East) game.
Hi, I tried to build your latest by following your wiki and end up having this error. But I tried your release v0.6.0 it works fine.
14:24:50.434253 src\main.rs:602 review has started, this may take several minutes...
14:24:50.444842 src\review.rs:219 reviewing kyoku=0 honba=0 junme=1 (0.17%)
Assertion failed: tehai_bit.count_hai(hai-5) >= tehai_state.get_aka_in_side(hai/10 - 1), file tehai_ana.cpp, line 263
Error: failed to review log
Caused by:
failed to read from akochan: unexpected EOF
Can the Mahjong Soul downloadlogs script be modified to output a format that is compatible with NAGA's new custom game analysis function?
I took the JSON output from a Mahjong Soul replay using these constants:
const NAMEPREF = 0; //2 for english, 1 for sane amount of weeb, 0 for japanese
const VERBOSELOG = false; //dump mjs records to output - will make the file too large for tenhou.net/5 viewer
const PRETTY = false; //make the written log somewhat human readable
and manually extracted each round to modify it to the below format using Notepad:
https://dl.dropboxusercontent.com/s/eslbmbxzmyv1zjv/nagaformat.txt?dl=0
After fixing the yaku name to 自風 東 in Round [7,0,0] to get past NAGA's yaku name validator (https://naga.dmv.nico/static/js/tenhou_edit.js), a successful analysis report was generated:
I'm trying this project recently,and have found some unexpected(not good) moves generated by akochan.For example,sometime (mostly when player is rank 3) it can't distinguish the diffirence between moves.
2m 3m 3m 5m 7p 1s 2s 3s 6s 6s 7s 8s 9s E
[6s] 期望[-55.90, -55.90, 0.002%, -137.63]
[3m]期望[-55.90, -55.90, 0.002%, -137.63]
[2s] 期望[-55.90, -55.90, 0.001%, -137.63]
[8s] 期望[-55.90, -55.90, 0.001%, -137.63]
[7s] 期望[-55.90, -55.90, 0.001%, -137.63]
[2m] 期望[-55.90, -55.90, 0.001%, -137.63]
[7p] 期望[-55.90, -55.90, 0.002%, -137.63]
[1s] 期望[-55.90, -55.90, 0.001%, -137.63]
[3s] 期望[-55.90, -55.90, 0.002%, -137.63]
[9s] 期望[-55.90, -55.90, 0.001%, -137.63]
[5m] 期望[-55.90, -55.90, 0.002%, -137.63]
[E] 期望[-55.90, -55.90, 0.000%, -137.63]
It seems that akochan's behavior is highly depended on parameters in "setup_mjai.json".Can you give me some advices on how to set these parameters to improve it? thank you very much! #
The Majsoul paipu is https://game.maj-soul.com/1/?paipu=220312-b4481a3d-57c0-4d89-b096-d7dd9dd7d837_a19427114
I followed the instruction here and used tempermonkey to download the paipu above. I tried to use akochan-reviewer to analyse it, using the command .\akochan-reviewer.exe -i .\2022_3_12_Gold_Room_South.json -n 0.1 --pt 110,50,-10,-130 -a 3 -k S1
, but found the 2nd turn was skipped when I pon 7m.
Running this command:
docker run --rm akochan-reviewer:latest --no-open -i - -a 1 -o - > report.html
Results in this error:
src/main.rs:334 parsing tenhou log...
Error: failed to parse tenhou log
Caused by:
EOF while parsing a value at line 1 column 0
According to the readme, it should read from stdin in this case.
Online page failed to read any mahjong soul url. Local build says cause by: get mahjong soul log: 503 Service Unavailable. The server is fine for normal players, could be the recent event update?
Hi, I'm MacOS 10.13 user, and I guess I successfully build both of akochan-reviewer and akochan.
But, when I test it, I confront the following issue.
$./akochan-reviewer -e "https://tenhou.net/0/?log=2019050417gm-0029-0000-4f2a8622&tw=2"
02:20:15.116799 src/main.rs:506 converting to mjai events...
02:20:15.249825 src/main.rs:600 players: カレーセット, 二宮蘭子, ⓝSuphx, 渡部惠子
02:20:15.249899 src/main.rs:601 target: ⓝSuphx
02:20:15.249946 src/main.rs:602 review has started, this may take several minutes...
02:20:15.253346 src/review.rs:219 reviewing kyoku=0 honba=0 junme=0 (0.95%)
02:20:15.376649 src/review.rs:219 reviewing kyoku=0 honba=0 junme=0 (1.58%)
02:20:15.482680 src/review.rs:219 reviewing kyoku=0 honba=0 junme=1 (1.90%)
02:20:15.603575 src/review.rs:361 review entry created: Agree (0/0/1, 100.000)
02:20:15.603706 src/review.rs:219 reviewing kyoku=0 honba=0 junme=1 (2.85%)
02:20:15.717992 src/review.rs:219 reviewing kyoku=0 honba=0 junme=1 (3.48%)
02:20:15.850804 src/review.rs:219 reviewing kyoku=0 honba=0 junme=1 (4.11%)
02:20:15.996262 src/review.rs:219 reviewing kyoku=0 honba=0 junme=2 (4.43%)
02:20:16.137382 src/review.rs:361 review entry created: Agree (0/0/2, 100.000)
02:20:16.137500 src/review.rs:219 reviewing kyoku=0 honba=0 junme=2 (5.38%)
02:20:16.259466 src/review.rs:219 reviewing kyoku=0 honba=0 junme=2 (6.01%)
02:20:16.372626 src/review.rs:219 reviewing kyoku=0 honba=0 junme=2 (6.65%)
02:20:16.488840 src/review.rs:219 reviewing kyoku=0 honba=0 junme=3 (6.96%)
Error: failed to review log
Caused by:
failed to read from akochan: unexpected EOF
I use Rust to build the akochan-reviewer.
I don't know this is the right place to ask the solution..
ADD
I guess this is because there is some segmentation fault issue in Akochan, not reviewer
% ./akochan-reviewer -a 0 -i 01.json
17:17:12.312590 src/main.rs:506 converting to mjai events...
Error: failed to canonicalize akochan_dir path "akochan"
Caused by:
No such file or directory (os error 2)
Windows 10でリリースv0.6.0が雀魂牌譜に限り動作不能になりました。URLを入力してしばらく置いていくとHTTP 500というエラーコードが出てしまう。できれば修復作業お願いします。
Error HTTP 500 appears for Jantama logs using release v0.6.0 on Windows 10
./akochan-reviewer.exe -e --lang en "http://tenhou.net/0/?log=2021121000gm-0029-0000-43854cc9&tw=3&tw=3"
Akochan recommends skipping the victory.
Is this a bug or a feature?
C:\Users\fan\Desktop\阿卡\akochan-reviewer>cargo build --release
Compiling winapi v0.3.9
Compiling proc-macro2 v1.0.32
Compiling syn v1.0.82
Compiling memchr v2.4.1
error: linker link.exe
not found
akochan-reviewer version: v0.4.4 (5811e5f)
Replay: https://tenhou.net/0/?log=2020100914gm-0089-0000-c74614a0&tw=3
東二局 2巡
If 牌譜 of special situations are listed, development becomes easier.
For example, short 牌譜、Double ロン、暗槓、加槓、大明槓、no akochan review etc...)
There is no need to prepare it immediately, so I want you to share your 牌譜ID somewhere when you find it.
Double ロン 牌譜 was found in the public 牌譜s in Tenhou site
Is it possible to provide prebuild binary for all system?
I analyzed this replay https://tenhou.net/0/?log=2021082901gm-0089-0000-454eab9a&tw=1
Command line:
./akochan-reviewer.exe --lang en -e -n 0.005 "https://tenhou.net/0/?log=2021082901gm-0089-0000-454eab9a&tw=1"
In South 3-1, turn 8, I got this:
MajsoulPaipuAnalyzer is a software developed by @zyr17 . The paipus file that the software uses is different from tenhou. Using that file to get paipus file is much easier than installing TamperMankey plugin and download then manually.
I've seen a few times when for the first few jun when Akochan thinks an honor tile should be discarded, no analysis is given even if the player didn't discard said tile.
Here's an example from a Suphx game
akochan-reviewer.exe --no-open -n 0.4 --pt 75,30,0,-105 "https://tenhou.net/4/?log=2019031218gm-0029-0000-d756d03c&tw=1"
docker version doesn't work with:
Error: failed to download mahjong soul log 220329-ce1da608-fc05-4560-bcfa-ff0185f91f92_a829847388
Caused by:
get mahjong soul log: 503 Service Unavailable
This error message shows when I was trying to run akochan-reviewer on windows 10 system, both tenhou and majsoul logs.
I used the pre-release version 0.6.0 in https://github.com/Equim-chan/akochan-reviewer/releases.
the full logs are as follows:
It seems the download script is using the old domain of Majsoul CN https://majsoul.union-game.com/0/ rather than the new one https://game.maj-soul.com/1/
Akochan does it somehow, but I can't get it to work locally trying to do this:
curl "https://tenhou.net/5/mjlog2json.cgi?log=2022041620gm-0009-0000-ff59f844&tw=1&tw=1" -H 'Content-Type: application/json' --data-raw '{ "Referer": "https://tenhou.net/"}' > ~/test.html
The idea is that I want to be able to save logs and then feed them to akochan later with -i command instead of relying on logs still being available and using akochan's fetcher.
It is easier to understand that the image of the "tehai" is displayed than the sentence "自家が ? を引いた時——".
It is difficult because akochan-viewer needs to save the state of the "tehai", but I think it will be a very good change if possible.
I came across the problem to get the game log. Also, I can get access to access https://tensoul.herokuapp.com/ normally. Would you be so kind to help me find a solution to this problem.
Error: failed to download mahjong soul log https://game.maj-soul.com/1/?paipu=210816-2576a4ac-3c87-472b-bcf0-2d7371a6b3a5_a415362495
Caused by:
get mahjong soul log: 500 Connection Failed
Thank you for your time.
.\akochan-reviewer.exe --lang en https://mahjongsoul.game.yo-star.com/?paipu=........
Error: failed to download mahjong soul log ....
Caused by:
get mahjong soul log: 503 Service Unaviable
The share link works good when used normally. The game was a ranked south game 4 players
version: akochan-reviewer-v0.6.0-windows-x86_64
I got "failed to read from akochan: unexpected EOF" error on Windows10 with akochan-reviewer windows binary v.0.6.0.
I need your help.
Situation
> akochan-reviewer "https://tenhou.net/0/?log=2019050417gm-0029-0000-4f2a8622&tw=2"
22:32:02.690380 src\main.rs:507 converting to mjai events...
22:32:02.691797 src\main.rs:603 players: カレーセット, 二宮蘭子, ⓝSuphx, 渡部惠子
22:32:02.692318 src\main.rs:604 target: ⓝSuphx
22:32:02.693135 src\main.rs:605 review has started, this may take several minutes...
22:32:02.705276 src\review.rs:221 reviewing kyoku=0 honba=0 junme=0 (0.95%)
22:32:03.183592 src\review.rs:221 reviewing kyoku=0 honba=0 junme=0 (1.58%)
22:32:03.631826 src\review.rs:221 reviewing kyoku=0 honba=0 junme=1 (1.90%)
22:32:04.078816 src\review.rs:374 review entry created: Agree (0/0/1, 100.000)
22:32:04.079945 src\review.rs:221 reviewing kyoku=0 honba=0 junme=1 (2.85%)
22:32:04.532977 src\review.rs:221 reviewing kyoku=0 honba=0 junme=1 (3.48%)
22:32:04.984261 src\review.rs:221 reviewing kyoku=0 honba=0 junme=1 (4.11%)
22:32:05.436933 src\review.rs:221 reviewing kyoku=0 honba=0 junme=2 (4.43%)
22:32:05.881838 src\review.rs:374 review entry created: Agree (0/0/2, 100.000)
22:32:05.883035 src\review.rs:221 reviewing kyoku=0 honba=0 junme=2 (5.38%)
22:32:06.331310 src\review.rs:221 reviewing kyoku=0 honba=0 junme=2 (6.01%)
22:32:06.780102 src\review.rs:221 reviewing kyoku=0 honba=0 junme=2 (6.65%)
22:32:07.228143 src\review.rs:221 reviewing kyoku=0 honba=0 junme=3 (6.96%)
Assertion failed: tehai_bit.count_hai(hai-5) >= tehai_state.get_aka_in_side(hai/10 - 1), file tehai_ana.cpp, line 263
Assertion failed: tehai_bit.count_hai(hai-5) >= tehai_state.get_aka_in_side(hai/10 - 1), file tehai_ana.cpp, line 263
Error: failed to review log
Caused by:
failed to read from akochan: unexpected EOF
> akochan-reviewer "https://game.mahjongsoul.com/?paipu=190425-146daa2a-68c2-4b7e-a8d7-2b5c71b54b00_a458023588"
22:43:29.293640 src\main.rs:507 converting to mjai events...
22:43:29.295242 src\main.rs:603 players: jan, 吼姆拉, 二宮蘭子, おむつみなと
22:43:29.295651 src\main.rs:604 target: 二宮蘭子
22:43:29.295982 src\main.rs:605 review has started, this may take several minutes...
22:43:29.307962 src\review.rs:221 reviewing kyoku=0 honba=0 junme=0 (0.26%)
Assertion failed: tehai_bit.count_hai(hai-5) >= tehai_state.get_aka_in_side(hai/10 - 1), file tehai_ana.cpp, line 263
Assertion failed: tehai_bit.count_hai(hai-5) >= tehai_state.get_aka_in_side(hai/10 - 1), file tehai_ana.cpp, line 263
Error: failed to review log
Caused by:
failed to read from akochan: unexpected EOF
Akochan fails to recognize log file urls from mjs.
Example that fails with the currently online version: https://mahjongsoul.game.yo-star.com/?paipu=210827-cdcae222-ee95-45d2-92fa-90b3ff1a9cfe_a939312232
I run akochan-reviewer in Ubuntu and got this error:
system.exe: error while loading shared libraries: libai.so: cannot open shared object file: No such file or directory
Error: failed to review log
Caused by:
failed to read from akochan: unexpected EOF
I am maintainer of 雀魂牌譜屋. Since my server caches log data for recent matches (usually for 1 ~ 2 days), I think akochan can use these data to allow reviewing recent matches. What do you think? If you like this idea, I can message you an HTTP endpoint for getting log data. :)
seems like there are missing files in the repository
22:47:30.309285 src/main.rs:338 parsing tenhou log...
22:47:30.316264 src/main.rs:370 converting to mjai events...
Error: failed to canonicalize akochan_exe path "/home/incertia/akochan-reviewer/akochan/system.exe"
Caused by:
No such file or directory (os error 2)
and also this is only runnable on windows
I am a beginner with computer and very interested in this AI, so I have just gone through the tutor and already built akochan and akochan-reviewer under my home dir.
By the way, my system is macOS 12.2.1 and when I was building akochan using make statement, my terminal gave me a lot of warnings, I don't know if this is relevant.
The problem is when I run '~/akochan-reviewer/target/release/akochan-reviewer -I -a 0' under home dir, this error occurs.
18:16:18.042392 src/main.rs:506 converting to mjai events...
Error: failed to canonicalize tactics_config path "tactics.json"
Caused by:
No such file or directory (os error 2)
But I can find my "tactics.json file" under /akochan-reviewer./akochan-reviewer/target/release/akochan-reviewer -I -a 0' under another dir, it will give me another error.
And when I run '
18:43:20.950012 src/main.rs:506 converting to mjai events...
Error: failed to canonicalize akochan_dir path "akochan"
Caused by:
No such file or directory (os error 2)
Can anyone tell me why this happen and how can I run this AI successfully on my Mac.
While you can currently import a MajSoul game, the point distribution presets only has Tenhou options. It would be nice to have the option to choose MajSoul presets or at the very least to manually enter the point distribution.
Hello Equim chan,
I occasionally found that akochan could not handle 四槓散了流局 properly.
I was using cmd
akochan-reviewer --pt 60,15,0,-90 -n 0.02 "https://tenhou.net/0/?log=2021052820gm-0089-0000-ef9e065c&tw=3"
where the address is https://tenhou.net/0/?log=2021052820gm-0089-0000-ef9e065c&tw=3
四槓散了 occured at 東4局1本場
And Akochan printed that
Error: failed to review log
Caused by:
0: invalid state in event
1: unexpected event: Ryukyoku { deltas: Some([0, 0, 0, 0]) }
I been using akochan reviewer for quite some time now, and recently I noticed it failed to get certain game log while other log replay is working perfectly fine.
Example:
Error: failed to download mahjong soul log 210426-42300644-9361-4b8f-a98e-e31293ae006c_a86463819
Caused by:
get mahjong soul log: 500 Internal Server Error
https://game.maj-soul.com/1/?paipu=210426-42300644-9361-4b8f-a98e-e31293ae006c_a86463819
Thank you for your time.
Like Issue #39, I was trying to review a Majsoul log, but this happened with two different log files. Then a window pop-up "System.exe has stopped working"
22:29:32.971615 src\main.rs:506 converting to mjai events...
22:29:32.972468 src\main.rs:600 players: 大型犬, 夜室泥霸霸, Rhdimer, catcat
22:29:32.972665 src\main.rs:601 target: catcat
22:29:32.972911 src\main.rs:602 review has started, this may take several minutes...
22:29:33.414583 src\review.rs:238 reviewing kyoku=0 honba=0 junme=0 (0.31%)
22:29:33.904226 src\review.rs:238 reviewing kyoku=0 honba=0 junme=0 (0.51%)
22:29:34.381490 src\review.rs:238 reviewing kyoku=0 honba=0 junme=0 (0.72%)
22:29:34.860264 src\review.rs:238 reviewing kyoku=0 honba=0 junme=1 (0.82%)
22:29:35.339958 src\review.rs:400 review entry created: Agree (0/0/1, 100.000)
22:29:35.340575 src\review.rs:238 reviewing kyoku=0 honba=0 junme=1 (1.13%)
22:29:35.819750 src\review.rs:238 reviewing kyoku=0 honba=0 junme=1 (1.34%)
22:29:36.294420 src\review.rs:238 reviewing kyoku=0 honba=0 junme=1 (1.54%)
22:29:36.777391 src\review.rs:238 reviewing kyoku=0 honba=0 junme=2 (1.65%)
Assertion failed: tehai_bit.count_hai(hai-5) >= tehai_state.get_aka_in_side(hai/10 - 1), file tehai_ana.cpp, line 263
Assertion failed: tehai_bit.count_hai(hai-5) >= tehai_state.get_aka_in_side(hai/10 - 1), file tehai_ana.cpp, line 263
Akochan reviewer version: v0.4.4
How to reproduce error:
akochan-reviewer.exe https://tenhou.net/0/?log=2020102101gm-00c1-0000-b7d8700d&tw=2
Error message:
1: Failed to render 'report.html'
2: Function call 'pretty_round' failed
3: Variable `metadata.score` was used in a math operation but is not a number
I was running akochan-reviewer to review Mahjong Soul Logs by following your instruction (https://github.com/Equim-chan/akochan-reviewer/blob/master/mjsoul.adoc) on Windows and failed.
running log:
➤ ./akochan-reviewer.exe -i 2021_12_6_Gold_Room_South.json -a 2
23:12:36.474591 src\main.rs:506 converting to mjai events...
23:12:36.475222 src\main.rs:600 players: 风条语顺, 钻石水果糖, Retr0MKII, 〓のなめ〓
23:12:36.475253 src\main.rs:601 target: Retr0MKII
23:12:36.475279 src\main.rs:602 review has started, this may take several minutes...
23:12:36.555729 src\review.rs:219 reviewing kyoku=0 honba=0 junme=0 (0.32%)
23:12:36.613702 src\review.rs:219 reviewing kyoku=0 honba=0 junme=0 (0.54%)
23:12:36.662356 src\review.rs:219 reviewing kyoku=0 honba=0 junme=1 (0.65%)
23:12:36.709961 src\review.rs:361 review entry created: Agree (0/0/1, 100.000)
23:12:36.710043 src\review.rs:219 reviewing kyoku=0 honba=0 junme=1 (0.97%)
23:12:36.759089 src\review.rs:219 reviewing kyoku=0 honba=0 junme=1 (1.18%)
23:12:36.806939 src\review.rs:219 reviewing kyoku=0 honba=0 junme=1 (1.40%)
23:12:36.855841 src\review.rs:219 reviewing kyoku=0 honba=0 junme=2 (1.51%)
Error: failed to review logCaused by:
failed to read from akochan: unexpected EOF
I tried on both ver 0.7.1 and 0.6.0 and both failed, however I run it on another PC(Windows) and it works. Maybe I miss some enviroment or something on my first PC?
Thanks for your patience. Have a nice day!
雀魂7月28日更新后,现在读取牌谱会失败。
7月28日雀魂アップデート後、現在読み方エラー発生します。
Now reading Majsoul log leads error after Majsoul updated on 28th July.
it may help in the majsoul reviewer process.
こんにちは。日本語で失礼いたします。
じゃんたまの東風戦の牌譜をローカルでakochanに検討させているのですが、オーラスのpt期待値を見るに南場が残っている想定で検討されているように思いました。
東風戦の想定で検討させることは可能でしょうか?前提からして間違っていたらごめんなさい。
When I follow the docker build instruction using M1 mac, I got the following error:
$ docker build -t akochan-reviewer:latest .
[+] Building 1.2s (10/15)
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 37B 0.0s
=> [internal] load .dockerignore 0.0s
=> => transferring context: 34B 0.0s
=> [internal] load metadata for docker.io/library/rust:1.45.2 0.9s
=> [ 1/11] FROM docker.io/library/rust:1.45.2@sha256:582bedbe2d3b7ada087 0.0s
=> [internal] load build context 0.1s
=> => transferring context: 65.68kB 0.1s
=> CACHED [ 2/11] RUN set -ex && apt-get update && apt-get install -y 0.0s
=> CACHED [ 3/11] WORKDIR /akochan-reviewer 0.0s
=> CACHED [ 4/11] COPY akochan akochan 0.0s
=> CACHED [ 5/11] WORKDIR /akochan-reviewer/akochan/ai_src 0.0s
=> ERROR [ 6/11] RUN set -ex && make -f Makefile_Linux libai.so && c 0.2s
------
> [ 6/11] RUN set -ex && make -f Makefile_Linux libai.so && cp libai.so ../:
#8 0.188 + make -f Makefile_Linux libai.so
#8 0.196 g++ tehai_cal_work.cpp -c -g -MMD -MP -std=c++11 -O3 -fopenmp -pthread -fPIC -DNPROCS=8 -mcmodel=medium -pedantic -Wignored-qualifiers -Wreturn-type -Wmaybe-uninitialized -Wbool-compare -Wshadow -Wunused-but-set-variable -Wunused-variable -lboost_system -o obj/tehai_cal_work.o
#8 0.198 g++: error: unrecognized argument in option '-mcmodel=medium'
#8 0.198 g++: note: valid arguments to '-mcmodel=' are: large small tiny
#8 0.198 make: *** [Makefile_Linux:34: obj/tehai_cal_work.o] Error 1
------
executor failed running [/bin/sh -c set -ex && make -f Makefile_Linux libai.so && cp libai.so ../]: exit code: 2
and I change akochan-reviewer/akochan/ai_src/Makefile_Linux
line3, -mcmodel=medium
to -mcmodel=small
I got error:
docker build -t akochan-reviewer:latest .
[+] Building 3.4s (10/15)
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 37B 0.0s
=> [internal] load .dockerignore 0.0s
=> => transferring context: 34B 0.0s
=> [internal] load metadata for docker.io/library/rust:1.45.2 1.7s
=> [ 1/11] FROM docker.io/library/rust:1.45.2@sha256:582bedbe2d3b7ada087ba37cd22ab266f085e58f019201d95972d4413e1e4651 0.0s
=> [internal] load build context 0.1s
=> => transferring context: 67.48kB 0.0s
=> CACHED [ 2/11] RUN set -ex && apt-get update && apt-get install -y libboost-all-dev && apt-get clean && rm -rf /var 0.0s
=> CACHED [ 3/11] WORKDIR /akochan-reviewer 0.0s
=> [ 4/11] COPY akochan akochan 0.1s
=> [ 5/11] WORKDIR /akochan-reviewer/akochan/ai_src 0.0s
=> ERROR [ 6/11] RUN set -ex && make -f Makefile_Linux libai.so && cp libai.so ../ 1.3s
------
> [ 6/11] RUN set -ex && make -f Makefile_Linux libai.so && cp libai.so ../:
#10 0.201 + make -f Makefile_Linux libai.so
#10 0.209 g++ tehai_cal_work.cpp -c -g -MMD -MP -std=c++11 -O3 -fopenmp -pthread -fPIC -DNPROCS=8 -mcmodel=small -pedantic -Wignored-qualifiers -Wreturn-type -Wmaybe-uninitialized -Wbool-compare -Wshadow -Wunused-but-set-variable -Wunused-variable -lboost_system -o obj/tehai_cal_work.o
#10 1.139 In file included from tehai_cal_work.hpp:5,
#10 1.139 from tehai_cal_work.cpp:1:
#10 1.139 tehai_action.hpp:8:14: error: enumerator value '-1' is outside the range of underlying type 'char'
#10 1.139 AT_DAHAI = -1,
#10 1.139 ^
#10 1.139 In file included from tehai_cal_work.hpp:5,
#10 1.139 from tehai_cal_work.cpp:1:
#10 1.139 tehai_action.hpp:30:16: error: enumerator value '-100' is outside the range of underlying type 'char'
#10 1.139 AT_NULL = -100,
#10 1.139 ^~~
#10 1.295 make: *** [Makefile_Linux:34: obj/tehai_cal_work.o] Error 1
------
executor failed running [/bin/sh -c set -ex && make -f Makefile_Linux libai.so && cp libai.so ../]: exit code: 2
I guess this Makefile_Linux
is for x86 and docker on M1 Mac will pull rust:1.45.2 container for arm
Can we do anything to fix this?
I've made a mistake sorry >.<
Hi, I am new to mahjong and find akochan-reviewer is useful for learning. Thanks a lot for this project!
However, I found it is a litter hard to check the review result and the tenhou replaying at the same time. There is only limited space if I don't collapse the tenhou replay iframe.
I try to add a horizontal layout to place the tenhou replay to the left and the review result to the right side. Here are my changes https://github.com/JaySon-Huang/akochan-reviewer/tree/hori_style
I am thinking of merging it back to this repo. Would you mind accepting the layout change?
A preview for the report with horizontal layout:
chankan.json.html.zip
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.