repo_name
stringlengths
1
62
dataset
stringclasses
1 value
lang
stringclasses
11 values
pr_id
int64
1
20.1k
owner
stringlengths
2
34
reviewer
stringlengths
2
39
diff_hunk
stringlengths
15
262k
code_review_comment
stringlengths
1
99.6k
serl
github_2023
python
7
rail-berkeley
youliangtan
@@ -0,0 +1,13 @@ +import numpy as np +import gymnasium as gym + +from franka_env.envs.franka_env import FrankaEnv + + +class FrankaPegInsert(FrankaEnv): + def __init__(self, **kwargs): + super().__init__(**kwargs) + + def go_to_rest(self, jpos=False): + # TODO: Implement this + return NotImpl...
I noticed that go_to_rest is not impl here added this TODO. Not sure if this is giving an issue. Also I make the workspace boundary configurable in the config class. Best if we can test it on the robot to ensure everything still works fine.
serl
github_2023
others
6
rail-berkeley
youliangtan
@@ -10,5 +10,5 @@ python async_drq_sim.py \ --utd_ratio 4 \ --batch_size 256 \ --eval_period 2000 \ - --encoder_type resnet \ + --encoder_type resnet-pretrained \ --debug
one tip is to add in the bash script ```bash python async_drq_sim.py "$@" ``` so this can support additional args that will arise in the future ```bash bash run_actor.sh --other_args 99 ```
serl
github_2023
python
6
rail-berkeley
youliangtan
@@ -0,0 +1,399 @@ +#!/usr/bin/env python3 + +# NOTE: this requires jaxrl_m to be installed: +# https://github.com/rail-berkeley/jaxrl_minimal + +import time +from functools import partial +import jax +import jax.numpy as jnp +import numpy as np +import tqdm +from absl import app, flags +from flax.training import ...
i assume these are part of the Charles PR?
serl
github_2023
python
6
rail-berkeley
youliangtan
@@ -0,0 +1,65 @@ +import gymnasium as gym +from tqdm import tqdm +import numpy as np +import copy + +import franka_env + +from franka_env.envs.relative_env import RelativeFrame +from franka_env.envs.wrappers import ( + GripperCloseEnv, + SpacemouseIntervention, + Quat2EulerWrapper, +) + +from serl_launcher.wra...
is this the utility script? im okay for this to be part of the example. Just make it under `if __name__ == '__main__'`
serl
github_2023
others
6
rail-berkeley
youliangtan
@@ -0,0 +1,16 @@ +export XLA_PYTHON_CLIENT_PREALLOCATE=false && \ +export XLA_PYTHON_CLIENT_MEM_FRACTION=.2 && \ +python async_drq_randomized.py \ + --learner \ + --env FrankaRobotiqPegInsert-Vision-v0 \ + --exp_name=serl_dev_drq_rlpd10demos_peg_insert_random_resnet_097 \ + --seed 0 \ + --random_steps 10...
hard coded `checkpoint_path`
serl
github_2023
python
6
rail-berkeley
youliangtan
@@ -0,0 +1,399 @@ +#!/usr/bin/env python3 + +# NOTE: this requires jaxrl_m to be installed: +# https://github.com/rail-berkeley/jaxrl_minimal + +import time +from functools import partial +import jax +import jax.numpy as jnp +import numpy as np +import tqdm +from absl import app, flags +from flax.training import ...
since we are not providing multithreaded example, feel free to remove those that uses `tunnel` or docs that mention it. FYI, in the edgeml library, the term `tunnel` refers to the shared datastore when using multi threading.
serl
github_2023
python
6
rail-berkeley
youliangtan
@@ -0,0 +1,399 @@ +#!/usr/bin/env python3 + +# NOTE: this requires jaxrl_m to be installed: +# https://github.com/rail-berkeley/jaxrl_minimal + +import time +from functools import partial +import jax +import jax.numpy as jnp +import numpy as np +import tqdm +from absl import app, flags +from flax.training import ...
should we rename this term `utd_ratio`? As UTD means updates to data ratio, and this might be confusing as we are doing a different thing here.
serl
github_2023
python
4
rail-berkeley
youliangtan
@@ -151,12 +151,13 @@ def sample( obs_pixels, self._num_stack + 1, axis=0 ) obs_pixels = obs_pixels[indx - self._num_stack] + obs_pixels = obs_pixels.transpose((0, 4, 1, 2, 3))
Add comment on this transpose seq
serl
github_2023
python
4
rail-berkeley
youliangtan
@@ -0,0 +1,359 @@ +#!/usr/bin/env python3 + +# NOTE: this requires jaxrl_m to be installed: +# https://github.com/rail-berkeley/jaxrl_minimal + +import time +from functools import partial +import jax +import jax.numpy as jnp +import numpy as np +import tqdm +from absl import app, flags + +import gymnasium as gym ...
instead of having 2 args, we can just keep `replay_buffer` as a single arg, initiate the iterator within the `learner()` function. cleaner
serl
github_2023
python
4
rail-berkeley
youliangtan
@@ -0,0 +1,359 @@ +#!/usr/bin/env python3 + +# NOTE: this requires jaxrl_m to be installed: +# https://github.com/rail-berkeley/jaxrl_minimal + +import time +from functools import partial +import jax +import jax.numpy as jnp +import numpy as np +import tqdm +from absl import app, flags + +import gymnasium as gym ...
Curiously, is this `done` redundant? or do we have this as truncated?
serl
github_2023
python
4
rail-berkeley
youliangtan
@@ -0,0 +1,396 @@ +#!/usr/bin/env python3 + +# NOTE: this requires jaxrl_m to be installed: +# https://github.com/rail-berkeley/jaxrl_minimal + +import time +from functools import partial +import jax +import jax.numpy as jnp +import numpy as np +import tqdm +from absl import app, flags + +import gymnasium as gym ...
similar to above
serl
github_2023
python
4
rail-berkeley
youliangtan
@@ -0,0 +1,396 @@ +#!/usr/bin/env python3 + +# NOTE: this requires jaxrl_m to be installed: +# https://github.com/rail-berkeley/jaxrl_minimal + +import time +from functools import partial +import jax +import jax.numpy as jnp +import numpy as np +import tqdm +from absl import app, flags + +import gymnasium as gym ...
The `demo_buffer` can be added within the `learner()` function. Also, it is preferable to make `franka_lift_cube_image_20_trajs.pkl"` configurable
serl
github_2023
python
4
rail-berkeley
youliangtan
@@ -0,0 +1,359 @@ +#!/usr/bin/env python3 + +# NOTE: this requires jaxrl_m to be installed: +# https://github.com/rail-berkeley/jaxrl_minimal + +import time +from functools import partial +import jax +import jax.numpy as jnp +import numpy as np +import tqdm +from absl import app, flags + +import gymnasium as gym ...
i think with the change of `learner()` args, this multithreaded option is no longer working. either we remove this, or support this. same to other examples.
serl
github_2023
python
3
rail-berkeley
youliangtan
@@ -0,0 +1,19 @@ +from setuptools import setup + +setup( + name="edgeml", + version="0.1.2",
`serl_launcher` package
serl
github_2023
python
3
rail-berkeley
youliangtan
@@ -0,0 +1,19 @@ +from setuptools import setup + +setup( + name="edgeml", + version="0.1.2", + description="library to enable distributed edge ml training and inference", + url="https://github.com/youliangtan/edgeml", + author="auth", + author_email="tan_you_liang@hotmail.com", + license="MIT", + ...
Since Iam maintaining `edgeml`, so it is better not to duplicate code ```py install_requires=[ ....... TODO DEPENDENCIES "'edgeml @ git+https://github.com/youliangtan/edgeml.git@COMMIT_HASH', ], ```
serl
github_2023
others
3
rail-berkeley
youliangtan
@@ -3,5 +3,36 @@ ![](https://github.com/rail-berkeley/serl/workflows/pre-commit/badge.svg) ## Installation -- Conda Environment: +1. Conda Environment: - create an environment with `conda create -n serl python=3.10` + +2. Install RL library + - the examples here use jaxrl-minimal as the RL library. + - T...
I see no diff in `serl_dev` and `main` in `jaxrl_minimal`, is that what it suppose to?
serl
github_2023
python
3
rail-berkeley
youliangtan
@@ -0,0 +1,195 @@ +from functools import partial
This seems to be similar to the one in `jaxrl_m`. If there's diff, place it in `serl_dev`, or keep it here for now (both options are good)
serl
github_2023
python
3
rail-berkeley
youliangtan
@@ -0,0 +1,162 @@ +import copy +from typing import Iterable, Optional, Tuple + +import gymnasium as gym +import numpy as np +from edgeml.data.serl_dataset import DatasetDict, _sample +from edgeml.data.serl_replay_buffer import ReplayBuffer +from flax.core import frozen_dict +from gymnasium.spaces import Box + +
Since this package is now call `serl_launcher`, the path should be: ```bash serl_launcher/serl_launcher/data/memory_efficient_replay_buffer.py ```
serl
github_2023
others
1
rail-berkeley
youliangtan
@@ -0,0 +1,5 @@ +# SERL: A Software Suite for Sample-Efficient Robotic Reinforcement Learning + +## Installation +- Conda Environment: + - create an environment with `conda create -n serl_dev python=3.10`
use `serl` as conda env name
bridge_data_v2
github_2023
others
9
rail-berkeley
HomerW
@@ -36,26 +36,46 @@ python experiments/train.py \ --name NAME ``` + Training hyperparameters can be modified in `experiments/configs/data_config.py` and data parameters (e.g. subsets to include/exclude) can be modified in `experiments/configs/train_config.py`. ## Evaluation -First, set up the robot hardw...
```suggestion To evaluate image-conditioned or language-conditioned methods, run `eval_gc.py` or `eval_lc.py` respectively in the docker container. (Refer to the [bridge_data_robot](https://github.com/rail-berkeley/bridge_data_robot) docs) ```
bridge_data_v2
github_2023
others
9
rail-berkeley
HomerW
@@ -36,26 +36,46 @@ python experiments/train.py \ --name NAME ``` + Training hyperparameters can be modified in `experiments/configs/data_config.py` and data parameters (e.g. subsets to include/exclude) can be modified in `experiments/configs/train_config.py`. ## Evaluation -First, set up the robot hardw...
```suggestion Optionally, you can run the `eval.py` which uses our newer `WidowXClient` API. With this setup, the robot is run as a server that receives actions and the policy acts as a client that sends actions. This "server-client" architecture allows us to both isolate robot controller and policy dependencies, as w...
bridge_data_v2
github_2023
others
9
rail-berkeley
HomerW
@@ -36,26 +36,46 @@ python experiments/train.py \ --name NAME ``` + Training hyperparameters can be modified in `experiments/configs/data_config.py` and data parameters (e.g. subsets to include/exclude) can be modified in `experiments/configs/train_config.py`. ## Evaluation -First, set up the robot hardw...
```suggestion export CHECKPOINT_DIR=/path/to/checkpoint_dir ```
bridge_data_v2
github_2023
others
9
rail-berkeley
HomerW
@@ -36,26 +36,46 @@ python experiments/train.py \ --name NAME ``` + Training hyperparameters can be modified in `experiments/configs/data_config.py` and data parameters (e.g. subsets to include/exclude) can be modified in `experiments/configs/train_config.py`. ## Evaluation -First, set up the robot hardw...
```suggestion python experiments/eval.py \ ```
bridge_data_v2
github_2023
others
9
rail-berkeley
HomerW
@@ -36,26 +36,46 @@ python experiments/train.py \ --name NAME ``` + Training hyperparameters can be modified in `experiments/configs/data_config.py` and data parameters (e.g. subsets to include/exclude) can be modified in `experiments/configs/train_config.py`. ## Evaluation -First, set up the robot hardw...
```suggestion python experiments/eval.py \ ```
bridge_data_v2
github_2023
others
9
rail-berkeley
HomerW
@@ -36,26 +36,46 @@ python experiments/train.py \ --name NAME ``` + Training hyperparameters can be modified in `experiments/configs/data_config.py` and data parameters (e.g. subsets to include/exclude) can be modified in `experiments/configs/train_config.py`. ## Evaluation -First, set up the robot hardw...
```suggestion --im_size 256 --goal_type gc --show_image --blocking ```
bridge_data_v2
github_2023
others
9
rail-berkeley
HomerW
@@ -36,26 +36,46 @@ python experiments/train.py \ --name NAME ``` + Training hyperparameters can be modified in `experiments/configs/data_config.py` and data parameters (e.g. subsets to include/exclude) can be modified in `experiments/configs/train_config.py`. ## Evaluation -First, set up the robot hardw...
```suggestion --im_size 256 --goal_type lc --show_image --blocking ```
bridge_data_v2
github_2023
python
9
rail-berkeley
HomerW
@@ -0,0 +1,385 @@ +#!/usr/bin/env python3 + +import sys +import os +import time +from datetime import datetime +import traceback +from collections import deque +import json + +from absl import app, flags, logging + +import numpy as np +import tensorflow as tf + +import cv2 +import jax +from PIL import Image +import ima...
Can you convert `obs["full_image"]` to BGR channel order here so that it's displayed correctly? (open-cv uses BGR channel order while our images are RGB)
bridge_data_v2
github_2023
python
9
rail-berkeley
HomerW
@@ -0,0 +1,385 @@ +#!/usr/bin/env python3 + +import sys +import os +import time +from datetime import datetime +import traceback +from collections import deque +import json + +from absl import app, flags, logging + +import numpy as np +import tensorflow as tf + +import cv2 +import jax +from PIL import Image +import ima...
Can we make `eval_lc.py` and `eval_gc.py` also have default `goal_eep` and default `initial_eep` of `[0.3, 0.0, 0.15]`? Though maybe not too important if we plan to deprecate them anyway.
smods
github_2023
others
580
Steamodded
english5040
@@ -152,12 +152,12 @@ function SMODS.calculate_retriggers(card, context, _ret) end ---@return table[] reps function SMODS.calculate_repetitions(card, context, reps) end +---@param blueprint_card Card|table ---@param card Card|table ----@param blueprint_card Card|table ---@param context CalcContext|table ---@ret...
I don't really like the original description- something like "Helper function to copy the ability of another joker. Useful for implementing Blueprint-like jokers" would be more clear. The current description makes it sound like this function is a calculation step, which it isn't.
smods
github_2023
others
428
Steamodded
Eremel
@@ -2261,81 +2298,199 @@ Set `prefix_config.key = false` on your object instead.]]):format(obj.key), obj. ----- API CODE GameObject.DeckSkin ------------------------------------------------------------------------------------------------- - local deck_skin_count_by_suit = {} SMODS.DeckSkins = {} - ...
Needs some sort of default value otherwise palettes that are missing it become the default card textures,
smods
github_2023
others
428
Steamodded
Eremel
@@ -982,7 +982,78 @@ SMODS.find_mod = function(id) return ret end +local function bufferCardLimitForSmallDS(cards, scaleFactor) + local cardCount = #cards + if type(scaleFactor) ~= "number" or scaleFactor <= 0 then + sendWarnMessage("scaleFactor must be a positive number") + return cardCount...
Needs a `not skin.outdated` check
smods
github_2023
others
428
Steamodded
Eremel
@@ -982,7 +982,78 @@ SMODS.find_mod = function(id) return ret end +local function bufferCardLimitForSmallDS(cards, scaleFactor) + local cardCount = #cards + if type(scaleFactor) ~= "number" or scaleFactor <= 0 then + sendWarnMessage("scaleFactor must be a positive number") + return cardCount...
This function doesn't allow custom colours on High Contrast and Low Contrast skins.
smods
github_2023
others
428
Steamodded
Eremel
@@ -780,6 +791,40 @@ function G.UIDEF.deck_preview(args) return t end +function tally_sprite(pos, value, tooltip, suit) + local text_colour = G.C.BLACK + if type(value) == "table" and value[1].string==value[2].string then + text_colour = value[1].colour or G.C.WHITE + value = value[1].string + end + local decksk...
This line crashes when viewing the deck
smods
github_2023
others
480
Steamodded
WilsontheWolf
@@ -0,0 +1,446 @@ +SMODS.DrawSteps = {} +SMODS.DrawStep = SMODS.GameObject:extend { + obj_table = SMODS.DrawSteps, + obj_buffer = {}, + required_params = { + 'key', + 'order', + 'func', + }, + layers = { + card = true, + both = true, + }, + -- func = function(card...
Could we do the no_shadow and shadows on check earlier in this function to avoid doing some extra work we don't need to?
smods
github_2023
others
480
Steamodded
WilsontheWolf
@@ -0,0 +1,446 @@ +SMODS.DrawSteps = {} +SMODS.DrawStep = SMODS.GameObject:extend { + obj_table = SMODS.DrawSteps, + obj_buffer = {}, + required_params = { + 'key', + 'order', + 'func', + }, + layers = { + card = true, + both = true, + }, + -- func = function(card...
Should we make a property that allows other cards to be able to get this behviour?
smods
github_2023
others
480
Steamodded
WilsontheWolf
@@ -0,0 +1,446 @@ +SMODS.DrawSteps = {} +SMODS.DrawStep = SMODS.GameObject:extend { + obj_table = SMODS.DrawSteps, + obj_buffer = {}, + required_params = { + 'key', + 'order', + 'func', + }, + layers = { + card = true, + both = true, + }, + -- func = function(card...
Same as above for invisible joker
smods
github_2023
others
480
Steamodded
WilsontheWolf
@@ -0,0 +1,446 @@ +SMODS.DrawSteps = {} +SMODS.DrawStep = SMODS.GameObject:extend { + obj_table = SMODS.DrawSteps, + obj_buffer = {}, + required_params = { + 'key', + 'order', + 'func', + }, + layers = { + card = true, + both = true, + }, + -- func = function(card...
```suggestion local seal = G.P_SEALS[self.seal] or {} ``` I don't believe that is nessicary
smods
github_2023
others
480
Steamodded
WilsontheWolf
@@ -0,0 +1,446 @@ +SMODS.DrawSteps = {} +SMODS.DrawStep = SMODS.GameObject:extend { + obj_table = SMODS.DrawSteps, + obj_buffer = {}, + required_params = { + 'key', + 'order', + 'func', + }, + layers = { + card = true, + both = true, + }, + -- func = function(card...
Same as the others above although this one seems less likely to want to be used
smods
github_2023
others
480
Steamodded
WilsontheWolf
@@ -0,0 +1,446 @@ +SMODS.DrawSteps = {} +SMODS.DrawStep = SMODS.GameObject:extend { + obj_table = SMODS.DrawSteps, + obj_buffer = {}, + required_params = { + 'key', + 'order', + 'func', + }, + layers = { + card = true, + both = true, + }, + -- func = function(card...
Putting this here as it's where I thought of it but it might be nice to add a propert to the draw step to skip this step in certain conditions. Something like ```lua skip = { vortex = true, front = true, } ``` which would just be equivalent to putting the code in a big old ```lua if not self.vortex and sel...
smods
github_2023
others
480
Steamodded
WilsontheWolf
@@ -0,0 +1,446 @@ +SMODS.DrawSteps = {} +SMODS.DrawStep = SMODS.GameObject:extend { + obj_table = SMODS.DrawSteps, + obj_buffer = {}, + required_params = { + 'key', + 'order', + 'func', + }, + layers = { + card = true, + both = true, + }, + -- func = function(card...
Same as the other's
smods
github_2023
others
480
Steamodded
WilsontheWolf
@@ -0,0 +1,446 @@ +SMODS.DrawSteps = {} +SMODS.DrawStep = SMODS.GameObject:extend { + obj_table = SMODS.DrawSteps, + obj_buffer = {}, + required_params = { + 'key', + 'order', + 'func', + }, + layers = { + card = true, + both = true, + }, + -- func = function(card...
This code would probably be cleaner with a table to lookup in and then mods could also add their own values to the table to not be drawn here if they want their own step.
smods
github_2023
others
480
Steamodded
WilsontheWolf
@@ -0,0 +1,446 @@ +SMODS.DrawSteps = {} +SMODS.DrawStep = SMODS.GameObject:extend { + obj_table = SMODS.DrawSteps, + obj_buffer = {}, + required_params = { + 'key', + 'order', + 'func', + }, + layers = { + card = true, + both = true, + }, + -- func = function(card...
Why do these need 2 different priorities?
smods
github_2023
others
480
Steamodded
GauntletGames-2086
@@ -0,0 +1,453 @@ +SMODS.DrawSteps = {} +SMODS.DrawStep = SMODS.GameObject:extend { + obj_table = SMODS.DrawSteps, + obj_buffer = {}, + required_params = { + 'key', + 'order', + 'func', + }, + layers = { + card = true, + both = true, + }, + -- func = function(card...
This and the Hologram checks should be swapped, so in the edgecase of someone taking ownership of Hologram to change the soul sprite `soul_pos.draw` is respected.
smods
github_2023
others
399
Steamodded
srockw
@@ -0,0 +1,407 @@ +---@meta + +--- Util Classes + +--- Internal class referring args passed as `context` in a SMODS object's `calculate` function. +--- Not all arguments typed here are present in all contexts, see [Calculate Function](https://github.com/Steamodded/smods/wiki/calculate_functions#contexts) for details. ...
Should be `remove_playing_cards`
smods
github_2023
others
399
Steamodded
srockw
@@ -0,0 +1,439 @@ +---@meta + +--- Util Classes + +--- Internal class referring args passed as `context` in a SMODS object's `calculate` function. +--- Not all arguments typed here are present in all contexts, see [Calculate Function](https://github.com/Steamodded/smods/wiki/calculate_functions#contexts) for details. ...
is actually `consumeable`
smods
github_2023
others
355
Steamodded
MathIsFun0
@@ -908,6 +950,240 @@ SMODS.find_mod = function(id) return ret end + + +-- This function handles the calculation of each effect returned to evaluate play. +-- Can easily be hooked to add more calculation effects ala Talisman +SMODS.calculate_individual_effect = function(effect, scored_card, percent, key, amount...
This is where the actual retriggering is, so context.retrigger_joker should be what is set to true here.
smods
github_2023
others
285
Steamodded
Aurelius7309
@@ -0,0 +1,119 @@ +[manifest] +version = "1.0.0" +dump_lua = true +priority = 0 + +#========================================================# +# Choose any rank for custom deck and use provided atlas # +#========================================================# +[[patches]] +[patches.regex] +target = "functions/misc_fu...
Using `pairs` here causes the suits to be out of order, I'd use `ipairs(SMODS.Suit:obj_list(true))` to be fully consistent
smods
github_2023
others
285
Steamodded
Aurelius7309
@@ -0,0 +1,119 @@ +[manifest] +version = "1.0.0" +dump_lua = true +priority = 0 + +#========================================================# +# Choose any rank for custom deck and use provided atlas # +#========================================================# +[[patches]] +[patches.regex] +target = "functions/misc_fu...
There's a crash when selecting a custom suit with no deck skins registered for that suit (apparently due to some localization entries missing). Tested with Bunco
smods
github_2023
others
285
Steamodded
Aurelius7309
@@ -0,0 +1,119 @@ +[manifest] +version = "1.0.0" +dump_lua = true +priority = 0 + +#========================================================# +# Choose any rank for custom deck and use provided atlas # +#========================================================# +[[patches]] +[patches.regex] +target = "functions/misc_fu...
Steamodded wants to preserve the behavior of the game for aspects that other mods don't actively change. As such, I can't approve always putting all ranks on display. At the very least the vanilla collabs should only display the face cards they're changing. Maybe you could make use of the `ranks` table to respect that?
smods
github_2023
others
314
Steamodded
WilsontheWolf
@@ -55,7 +55,7 @@ function loadMods(modsDirectory) }, prefix = { pattern = '%-%-%- PREFIX: (.-)\n' }, version = { pattern = '%-%-%- VERSION: (.-)\n', handle = function(x) return x and V(x):is_valid() and x or '0.0.0' end }, - outdated = { pattern = { 'SMODS%.INIT', 'S...
the %. is nessicary as . represents any character, and the % escapes it
smods
github_2023
others
243
Steamodded
Aurelius7309
@@ -7,32 +7,47 @@ priority = 0 # Card:open [[patches]] -[patches.pattern] +[patches.regex] target = "card.lua" -pattern = '''G.GAME.pack_choices = self.config.center.config.choose or 1''' +pattern = '''(?<indent>[\t ]*)if self\.ability\.name:find\('Arcana'\) then \n[\s\S]{12}G\.STATE''' position = "before" paylo...
This causes no issues if done correctly, as the function can be hooked with access to the pre-existing version, changing parts of the return value as necessary. Conflicts, should they occur, are decided by loading priority.
smods
github_2023
others
206
Steamodded
Aurelius7309
@@ -0,0 +1,459 @@ +--- STEAMODDED HEADER +--- MOD_NAME: Vanilla SMODS +--- MOD_ID: VSMODS +--- MOD_AUTHOR: [someone23832] +--- MOD_DESCRIPTION: Turns some vanilla jokers into an SMODS format, and changes them to give some examples of what you can do. +--- LOADER_VERSION_GEQ: 1.0.0 +--- BADGE_COLOR: c7638f +--- PREFIX: ...
These functions are duplicates; `config.extra.d_size` and `h_size` apply these automatically. So this is actually oversaturated Merry Andy
smods
github_2023
others
217
Steamodded
english5040
@@ -1939,40 +1939,57 @@ Set `prefix_config.key = false` on your object instead.]]):format(obj.key), obj. ----- API CODE GameObject.PokerHand ------------------------------------------------------------------------------------------------- + SMODS.PokerHandParts = {} + SMODS.PokerHandPart = SMODS.GameO...
I support this, since it was confusing. Reminder that we should warn people about this
smods
github_2023
others
217
Steamodded
english5040
@@ -1939,40 +1939,57 @@ Set `prefix_config.key = false` on your object instead.]]):format(obj.key), obj. ----- API CODE GameObject.PokerHand ------------------------------------------------------------------------------------------------- + SMODS.PokerHandParts = {} + SMODS.PokerHandPart = SMODS.GameO...
An idea I had for poker hands: auto-sort them based on the product of base mult/chips; above_hand (if set) should be considered first. If two hands are both set to be above the same hand, there could also be conflicts?
smods
github_2023
others
217
Steamodded
english5040
@@ -1986,6 +2003,70 @@ Set `prefix_config.key = false` on your object instead.]]):format(obj.key), obj. inject = function(self) end } + SMODS.PokerHandPart { + key = '_highest', + func = function(hand) return get_highest(hand) end + } + SMODS.PokerHandPart { + key = '_strai...
A bit misleading to call this _fh if it's not necessarily a full house Edit: yeah, if it's called _fh people might use it when it's just a pair. something like all_pairs or something might be better
smods
github_2023
others
207
Steamodded
Aurelius7309
@@ -323,7 +324,7 @@ function buildAdditionsTab(mod) local modNodes = {} table.insert(modNodes, t) - return { + return (#leftside_nodes > 0 and #rightside_nodes > 0 ) and {
This doesn't seem right...? It's possible for one column to be empty and the tab should still show
smods
github_2023
others
207
Steamodded
Aurelius7309
@@ -2148,9 +2146,85 @@ Set `prefix_config.key = false` on your object instead.]]):format(obj.key), obj. SMODS.process_loc_text(G.localization.descriptions.Other, self.key, self.loc_txt, 'description') SMODS.process_loc_text(G.localization.misc.labels, self.key, self.loc_txt, 'label') ...
`sticker_check = nil` does nothing, it acts like there is no value there. Use 0 or false instead. Though the base checks should work here?
smods
github_2023
others
207
Steamodded
Aurelius7309
@@ -494,6 +527,157 @@ G.UIDEF.consumable_collection_page = function(page) return t end +G.FUNCS.your_collection_stickers = function(e) + G.SETTINGS.paused = true + G.FUNCS.overlay_menu{ + definition = create_UIBox_your_collection_stickers(), + } +end + +function create_UIBox_your_collection_stickers(exit) + loca...
Missing 'end' here
smods
github_2023
others
212
Steamodded
Aurelius7309
@@ -584,3 +584,43 @@ function format_ui_value(value) end --#endregion + + +function poll_seal(_key, _mod, _guaranteed, _options, _type_key)
Consistency, this should be `SMODS.poll_seal`. API design, this has too many arguments. I suggest converting to a table argument that can hold these arguments as needed
smods
github_2023
others
212
Steamodded
Aurelius7309
@@ -584,3 +584,43 @@ function format_ui_value(value) end --#endregion + + +function poll_seal(_key, _mod, _guaranteed, _options, _type_key) + _mod = _mod or 1 + local seal_poll = pseudorandom(pseudoseed(_key or 'stdseal'..G.GAME.round_resets.ante)) + -- If no _options provided, get all seal keys + if no...
In reference to #155, this should expose the same `get_weight` function that editions do.
smods
github_2023
others
212
Steamodded
Aurelius7309
@@ -9,15 +9,35 @@ priority = 0 [patches.regex] target = 'card.lua' pattern = ''' -(?<indent>[\t ]*)if seal_type > 0.75 then card:set_seal\('Red'\) -[\t ]*elseif seal_type > 0.5 then card:set_seal\('Blue'\) -[\t ]*elseif seal_type > 0.25 then card:set_seal\('Gold'\) -[\t ]*else card:set_seal\('Purple'\) -[\t ]*end'''...
Consistency: why does this exclude modded seals?
smods
github_2023
others
212
Steamodded
Aurelius7309
@@ -179,4 +199,32 @@ position = "after" payload = ''' elseif self.children.alert and self.seal and not G.P_SEALS[self.seal].alerted then G.P_SEALS[self.seal].alerted = true - G:save_progress()''' \ No newline at end of file + G:save_progress()''' + +# Game:init_item_prototyp...
Invalidates `order`, can we change the order on P_SEALS directly?
smods
github_2023
others
202
Steamodded
english5040
@@ -395,6 +395,21 @@ function modsCollectionTally(pool, set) return obj_tally end +function math.round(num, precision) + precision = 10^(precision or 0) + + return math.floor(num * precision + 0.49999999999999994) / precision
I advise not overwriting stuff in the math namespace (liable to lead to incompatibilities, misleading because it's not in standard library). If possible, is there a way you could make this more mathematically precise? I understand if you can't
smods
github_2023
others
202
Steamodded
english5040
@@ -0,0 +1,70 @@ +[manifest] +version = "1.0.0" +dump_lua = true +priority = 0 + + +# +# End of round money +# + +# Hide off screen rows + +[[patches]] +[patches.pattern] +target = "functions/common_events.lua" +pattern = "if config.name ~= 'bottom' then" +position = "at" +payload = ''' +if config.name ~= 'bottom' then...
round_eval_rows shouldn't be a global, it could lead to weirdness when saving/loading
smods
github_2023
others
202
Steamodded
english5040
@@ -395,6 +395,21 @@ function modsCollectionTally(pool, set) return obj_tally end +function math.round(num, precision) + precision = 10^(precision or 0) + + return math.floor(num * precision + 0.49999999999999994) / precision +end + +-- Formatting util for UI elements (look number_formatting.toml) +function forma...
move these functions to utils. though , file structure remains messy, sorry
smods
github_2023
others
204
Steamodded
english5040
@@ -1,83 +1,6 @@ --- STEAMODDED CORE --- MODULE MODLOADER --- Attempt to require nativefs -local nfs_success, nativefs = pcall(require, "nativefs") -local lovely_success, lovely = pcall(require, "lovely") - -local lovely_mod_dir -local library_load_fail = false -if lovely_success then - lovely_mod_dir = lovely.m...
Concerned about removing the block above.
smods
github_2023
others
151
Steamodded
english5040
@@ -2419,165 +2423,237 @@ function loadAPIs() }) ------------------------------------------------------------------------------------------------- - ----- API CODE GameObject.Palette + ----- API CODE GameObject.AltTexture ---------------------------------------------------------------------------...
The organization here is a bit weird, Types is not a 'real' element of the table
smods
github_2023
others
151
Steamodded
english5040
@@ -2419,165 +2423,237 @@ function loadAPIs() }) ------------------------------------------------------------------------------------------------- - ----- API CODE GameObject.Palette + ----- API CODE GameObject.AltTexture ---------------------------------------------------------------------------...
Just use {} here, if the SMODS.alt_textures name is not intended to be used
smods
github_2023
others
151
Steamodded
english5040
@@ -2419,165 +2423,237 @@ function loadAPIs() }) ------------------------------------------------------------------------------------------------- - ----- API CODE GameObject.Palette + ----- API CODE GameObject.AltTexture ---------------------------------------------------------------------------...
from_palette, from_atlas seem more natural. after all, we aren't creating a palette, we're creating a texture from a palette
smods
github_2023
others
151
Steamodded
english5040
@@ -1017,87 +1016,87 @@ end --#endregion --#region enhancements UI function create_UIBox_your_collection_enhancements(exit) - local deck_tables = {} - local rows, cols = 2, 4 - local page = 0 - - G.your_collection = {} - for j = 1, rows do - G.your_collection[j] = CardArea(G.ROOM.T.x + 0.2 * G.ROOM.T.w / 2, G.ROOM....
Sort Enhancements when they're initially defined, not here
smods
github_2023
others
151
Steamodded
english5040
@@ -388,73 +388,93 @@ function serialize_string(s) return string.format("%q", s) end ---#region palettes -G.SETTINGS.selected_colours = G.SETTINGS.selected_colours or {} +--#region alt textures +G.SETTINGS.selected_texture = G.SETTINGS.selected_texture or {} G.PALETTE = {} +G.C.AltTexture = HEX("277a97") +default...
slight error in comment: palette -> 'texture'
smods
github_2023
others
151
Steamodded
english5040
@@ -388,73 +388,93 @@ function serialize_string(s) return string.format("%q", s) end ---#region palettes -G.SETTINGS.selected_colours = G.SETTINGS.selected_colours or {} +--#region alt textures +G.SETTINGS.selected_texture = G.SETTINGS.selected_texture or {} G.PALETTE = {}
is this needed?
smods
github_2023
others
151
Steamodded
english5040
@@ -388,73 +388,93 @@ function serialize_string(s) return string.format("%q", s) end ---#region palettes -G.SETTINGS.selected_colours = G.SETTINGS.selected_colours or {} +--#region alt textures +G.SETTINGS.selected_texture = G.SETTINGS.selected_texture or {} G.PALETTE = {} +G.C.AltTexture = HEX("277a97") +default...
Not sure what this is doing now
smods
github_2023
others
151
Steamodded
english5040
@@ -805,22 +809,16 @@ function loadAPIs() SMODS.process_loc_text(G.localization.descriptions.Other, 'undiscovered_' .. string.lower(self.key), self.loc_txt, 'undiscovered') end, - generate_colours = function(self, base_colour, alternate_colour) - if not self.colo...
Concerned about adding this- I don't want to add things to SMODS that are not generally useful utilities, this seems pretty specific.
smods
github_2023
others
151
Steamodded
english5040
@@ -2302,10 +2310,6 @@ function loadAPIs() SMODS.process_loc_text(G.localization.misc.labels, self.key:sub(3), self.loc_txt, 'label') SMODS.Edition.super.process_loc_text(self) end, - -- apply_modifier = true when G.GAME.edition_rate is to be applied - get_weight = funct...
Why was get_weight removed? also get_weight remains a bad name, sorry. to be honest, letting weight itself be either a number or a function might be better
smods
github_2023
others
151
Steamodded
english5040
@@ -2419,165 +2423,237 @@ function loadAPIs() }) ------------------------------------------------------------------------------------------------- - ----- API CODE GameObject.Palette + ----- API CODE GameObject.AltTexture ---------------------------------------------------------------------------...
Again, create_colours is something I am wary of including in Steamodded itself. I'd like these functions to be cohesive, any specific functions can go in a separate mod.
smods
github_2023
others
151
Steamodded
Aurelius7309
@@ -2419,165 +2423,237 @@ function loadAPIs() }) ------------------------------------------------------------------------------------------------- - ----- API CODE GameObject.Palette + ----- API CODE GameObject.AltTexture ---------------------------------------------------------------------------...
These `take_ownership` calls flood logs at startup for no good reason. Perhaps adding a way to make them not log anything would be reasonable
smods
github_2023
others
151
Steamodded
english5040
@@ -102,16 +102,12 @@ function loadAPIs() -- Setup Localize text o:process_loc_text() - - sendInfoMessage( - ('Registered game object %s of type %s') - :format(o.key, o.set), o.set or 'GameObject' - ) + if not o.no_log then sendI...
for the no-logging functionality, I was thinking of a setting on the logging system, not a setting per-object. basically, some toggle that temporarily prevents logging. It could be as simple as a global variable- that's not extensible, but it's also simple enough to be fine
smods
github_2023
others
138
Steamodded
Aurelius7309
@@ -22,14 +22,11 @@ function loadAPIs() function SMODS.GameObject:__call(o) o = o or {} o.mod = SMODS.current_mod - if o.mod and not o.raw_atlas_key and not (o.mod.omit_mod_prefix or o.omit_mod_prefix) then + if o.mod and not o.raw_atlas_key and not (o.mod.omit_mod_prefix or o.omit_...
Why should a `shader` being present prevent atlas keys from being prefixed?
smods
github_2023
python
10
Steamodded
Steamopollys
@@ -85,127 +92,145 @@ def modify_game_lua(game_lua_path): print("Successfully modified game.lua.") else: print("Target line not found in game.lua.") + raise RuntimeError() except IOError as e: print(f"Error modifying game.lua: {e}") + raise RuntimeErr...
`base_dir = os.path.dirname(os.path.abspath(__file__))` To match new file structure
smods
github_2023
python
10
Steamodded
Steamopollys
@@ -85,127 +92,145 @@ def modify_game_lua(game_lua_path): print("Successfully modified game.lua.") else: print("Target line not found in game.lua.") + raise RuntimeError() except IOError as e: print(f"Error modifying game.lua: {e}") + raise RuntimeErr...
`# Running in a PyInstaller bundle`
smods
github_2023
python
10
Steamodded
Steamopollys
@@ -85,127 +92,145 @@ def modify_game_lua(game_lua_path): print("Successfully modified game.lua.") else: print("Target line not found in game.lua.") + raise RuntimeError() except IOError as e: print(f"Error modifying game.lua: {e}") + raise RuntimeErr...
`# Running in a normal Python environment or Nuitka bundle`
smods
github_2023
others
43
Steamodded
jubnl
@@ -73,6 +72,9 @@ function SMODS.injectJokers() effect = joker.effect, cost = joker.cost, cost_mult = 1.0, + blueprint_compat = joker.blueprint_compat, + eternal_compat = joker.eternal_compat, + effect = joker.effect
The line 77 does not seems necessary seems the joker effect is already extracted at line 72. The line 77 also crash the game becase there's a missing comma at the end of the line
smods
github_2023
python
1
Steamodded
Steamopollys
@@ -76,13 +76,13 @@ def modify_main_lua(main_lua_path, base_dir, directories): # print("LuaJIT Decompiler downloaded.") # URL to download the 7-Zip suite - seven_zip_url = "https://cdn.discordapp.com/attachments/485484159603572757/1181530870029504514/7-Zip.zip?ex=6581655f&is=656ef05f&hm=3dfd3e5a4936b0a50...
7z2301-x64.exe
smods
github_2023
python
1
Steamodded
Steamopollys
@@ -76,13 +76,13 @@ def modify_main_lua(main_lua_path, base_dir, directories): # print("LuaJIT Decompiler downloaded.") # URL to download the 7-Zip suite - seven_zip_url = "https://cdn.discordapp.com/attachments/485484159603572757/1181530870029504514/7-Zip.zip?ex=6581655f&is=656ef05f&hm=3dfd3e5a4936b0a50...
7z2301-x64.exe
smods
github_2023
others
589
Steamodded
GauntletGames-2086
@@ -866,6 +866,11 @@ Set `prefix_config.key = false` on your object instead.]]):format(obj.key), obj. prefix_config = { key = false }, inject = function(self) G.P_CENTER_POOLS[self.key] = G.P_CENTER_POOLS[self.key] or {} + if self.cards then + for key, _ in pairs...
The game will crash if any of the keys within `cards` are not vanilla (since they're not injected until after `SMODS.ObjectType` is.
git-backdate
github_2023
others
10
rixx
rixx
@@ -122,7 +133,7 @@ def rewrite_history( ], env=dict(os.environ, GIT_COMMITTER_DATE=timestamp.isoformat()), ) - last_timestamp = timestamp + prev_timestamp = timestamp
So this is where I'm really baffled – not only is `last_timestamp` at least as correct as `prev_timestamp`, it's also two full words rather than an abbreviation. Matter of taste, I suppose, but I don't see how it's relevant to this PR.
git-backdate
github_2023
others
10
rixx
rixx
@@ -82,32 +91,34 @@ def rewrite_history( elif no_business_hours: min_hour = 18 max_hour = 23 - duration = len(days) - commits_per_day = math.ceil(len(commits) / duration) - last_timestamp = None - commit_count = len(commits) - day_progress = 0 - for commit_index in range(commit_...
Really sorry to be grumpy, but this place illustrates beautifully how short and readable this PR could have been without the renaming of variables.
git-backdate
github_2023
others
10
rixx
rixx
@@ -82,32 +91,34 @@ def rewrite_history( elif no_business_hours: min_hour = 18 max_hour = 23 - duration = len(days)
As `len(days)` doesn't change, isn't it preferable to compute it once outside the loop rather than again and again inside the loop?
git-backdate
github_2023
others
10
rixx
rixx
@@ -82,32 +91,34 @@ def rewrite_history( elif no_business_hours: min_hour = 18 max_hour = 23 - duration = len(days) - commits_per_day = math.ceil(len(commits) / duration) - last_timestamp = None - commit_count = len(commits) - day_progress = 0 - for commit_index in range(commit_...
This expression can throw an error, as `prev_timestamp.hour` is not guaranteed to be set here.
git-backdate
github_2023
others
10
rixx
rixx
@@ -82,32 +91,34 @@ def rewrite_history( elif no_business_hours: min_hour = 18 max_hour = 23 - duration = len(days) - commits_per_day = math.ceil(len(commits) / duration) - last_timestamp = None - commit_count = len(commits) - day_progress = 0 - for commit_index in range(commit_...
If the previous timestamp is significantly off our start, this will screw over all preceding calculations re: commits per day.
diagram
github_2023
others
38
pandoc-ext
tarleb
@@ -80,6 +80,7 @@ that can be used to specify a specific executable. | [Mermaid] | `mermaid` | `mmdc` | `MERMAID_BIN` | | [PlantUML] | `plantuml` | `plantuml` | `PLANTUML_BIN` | | [Ti*k*Z] | `tikz` | `pdflatex` | `PDFLATEX_BIN` | +| [cetz] | `cetz` | `typst` ...
The package itself seems to use the capitalization `CeTZ`.
mrjs
github_2023
javascript
599
Volumetrics-io
michaelthatsit
@@ -101,24 +103,95 @@ export class AnimationSystem extends MRSystem { /** * @function - * @description Sets the Animation of the entity object based on the component value associated with it. + * @description Sets the Animation of the entity object based on the component value associated with it. O...
You can also do `comp?.action` which is a little cleaner.
mrjs
github_2023
javascript
599
Volumetrics-io
michaelthatsit
@@ -101,24 +103,95 @@ export class AnimationSystem extends MRSystem { /** * @function - * @description Sets the Animation of the entity object based on the component value associated with it. + * @description Sets the Animation of the entity object based on the component value associated with it. O...
Maybe 'repeat' instead of 'loopMode'?
mrjs
github_2023
javascript
578
Volumetrics-io
michaelthatsit
@@ -12,56 +12,65 @@ import { MRTextInputEntity } from 'mrjs/core/entities/MRTextInputEntity'; export class MRTextAreaEntity extends MRTextInputEntity { /** * @class + * @description Constructor for the MRTextInputEntity entity component. */ constructor() { super(); - // Define...
I think we talked about this but why aren't we putting this in the shadow root? might scale a little better. especially for embedded experiences.
mrjs
github_2023
javascript
578
Volumetrics-io
michaelthatsit
@@ -16,55 +14,91 @@ export class MRTextFieldEntity extends MRTextInputEntity { */ constructor() { super(); - this.wrapper = this.shadowRoot.appendChild(document.createElement('div')); - this.wrapper.innerHTML = '<slot></slot>'; + this.lineHeight = 1.2; // Default line height, ca...
Related to my other comment. why a hidden field in the DOM instead of an actual field in the shadow?
mrjs
github_2023
javascript
578
Volumetrics-io
michaelthatsit
@@ -73,7 +107,7 @@ export class MRTextFieldEntity extends MRTextInputEntity { // Ensure the cursor position is updated to reflect the current caret position setTimeout(() => {
what's this for?
mrjs
github_2023
javascript
581
Volumetrics-io
hanbollar
@@ -116,7 +120,13 @@ export class PhysicsSystem extends MRSystem { if (entity instanceof MRDivEntity) { this.initPhysicsBody(entity); this.registry.add(entity); - entity.addEventListener('modelchange', this.entityEventUpdate); + entity.addEventListener('modelchan...
if child-updated is one of our own events, it would be better as `childupdated` since we're switching away from the hyphenated ones
mrjs
github_2023
javascript
581
Volumetrics-io
hanbollar
@@ -75,7 +75,7 @@ export class MRImageEntity extends MRMediaEntity { // passed through MRMediaEntity directly is undefined since it is not // a direct element for users. So we do the if-check here and then // follow the same as the parent's functionality. - if (mutation.type != 'attrib...
not sure i understand the reasoning for this swap - was this originally a typo?
mrjs
github_2023
javascript
581
Volumetrics-io
hanbollar
@@ -137,7 +137,7 @@ export class MRMediaEntity extends MRDivEntity { mutated(mutation) { super.mutated(); - if (mutation.type != 'attributes' && mutation.attributeName == 'src') { + if (mutation.type == 'attributes' && mutation.attributeName == 'src') {
same question for here