silksong-autosplit-wasm

KO | EN

gitlite — search

#javascript #python #ai #typescript #llm #android #machine-learning #deep-learning #linux #macos #cli #kubernetes #vue #security

silksong-autosplit-wasm

(★ 19)

An auto splitter for Hollow Knight: Silksong

silksong-autosplit-wasm 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# silksong-autosplit-wasm

An auto splitter for Hollow Knight: Silksong.

## Installation

### LiveSplit (Windows)

Get [LiveSplit](https://livesplit.org/downloads/), 1.8.34 or later.

Right-click for the context menu:
- Open Splits, From File... : Select your `.lss` splits file. Go to [HKSplitMaker](https://hksplitmaker.com/?game=silksong) to generate and download `.lss` splits files.
- Edit Splits... : Activate the autosplitter.
- Compare Against: Game Time.

See also:
- [Silksong-Resources LiveSplit Setup Guide](https://github.com/hk-speedrunning/Silksong-Resources/blob/main/setup.md#livesplit)

### LiveSplit One Druid (Windows, Linux, Mac)

Go to the [LiveSplit One Druid Latest Release](https://github.com/AlexKnauth/livesplit-one-druid/releases/latest) page,
and under the `Assets` section, download the one for your architecture and operating system.

When you run LiveSplitOne, it needs to have permission to read memory of other processes.
- Windows: no additional steps required.
- Linux: set the capabilities to include `CAP_SYS_PTRACE`, with a command like `sudo setcap CAP_SYS_PTRACE=+eip LiveSplitOne` to run once after downloading LiveSplitOne.
- Mac: you have to run it under `sudo`, with a command like `sudo ./LiveSplitOne` to run every time you want to open it.

Right-click or Control-click for the context menu:
- Splits, Open... : Select your `.lss` splits file. Go to [HKSplitMaker](https://hksplitmaker.com/?game=silksong) to generate and download `.lss` splits files.
- Open Auto-splitter... : Select the `silksong_autosplit_wasm_stable.wasm` file. Go to the [silksong-autosplit-wasm Latest Release](https://github.com/AlexKnauth/silksong-autosplit-wasm/releases/latest) to download that.
- Compare Against: Game Time.
- Hotkeys: Configure the hotkeys you want. The default hotkeys use numpad, so if your computer doesn't have a numpad, configure them differently.

### OBS LiveSplit One (Windows, Linux)

Go to the [OBS LiveSplit One Latest Release](https://github.com/AlexKnauth/obs-livesplit-one/releases/latest) page,
and under the `Assets` section, download the one for your architecture and operating system.
Follow the instructions in [How to install](https://github.com/AlexKnauth/obs-livesplit-one?tab=readme-ov-file#how-to-install):
- Windows: Extract the `obs-livesplit-one.dll` to `C:\Program Files\obs-studio\obs-plugins\64bit` or equivalent install directory.
- Linux: Ensure the plugins folder exists with `mkdir -p $HOME/.config/obs-studio/plugins`, then extract with a command like `tar -zxvf obs-livesplit-one-*-x86_64-unknown-linux-gnu.tar.gz -C $HOME/.config/obs-studio/plugins/`.

When you run OBS, it needs to have permission to read memory of other processes.
- Windows: no additional steps required.
- Linux: set the capabilities to include `CAP_SYS_PTRACE`, with a command like `sudo setcap CAP_SYS_PTRACE=+eip /usr/bin/obs` to run once after downloading OBS.

Add OBS Source: LiveSplit One.

Properties:
- Splits: Select your splits file. Go to [HKSplitMaker](https://hksplitmaker.com/?game=silksong) to generate and download `.lss` splits files.
- Activate

Open the OBS Settings from File, Settings:
- Go to the Hotkeys section and scroll down until you find LiveSplit One.
- Set a hotkey for `Toggle Timing Method`, and hit Ok.
- Hit that hotkey once to switch from the default, Real Time, to Game Time.

## Custom Variables: percent, hits, deaths

You can show the completion percent with Edit layout:
- Plus, Information, Text
- Layout settings, Text:
  - check the box for Custom Variable
  - Custom Variable Name: `percent`
 
Some percents might not update the `percent` immediately, but will update on the next save point or when you open inventory.
 
If you have the Hit Counter setting turned on, you can show the number of hits with Edit Layout:
- Plus, Information, Text
- Layout settings, Text:
  - check the box for Custom Variable
  - Custom Variable Name: `hits`
 
You can also send hits to HitCounterManager via the [LiveSplit.HitCounterManagerConnector](https://github.com/topeterk/LiveSplit.HitCounterManagerConnector) component.

If you have the Death Counter setting turned on, you can show the number of deaths with Edit Layout:
- Plus, Information, Text
- Layout settings, Text:
  - check the box for Custom Variable
  - Custom Variable Name: `deaths`

## Compilation

This auto splitter is written in Rust. In order to compile it, you need to
install the Rust compiler: [Install Rust](https://www.rust-lang.org/tools/install).

Afterwards install the WebAssembly target:
```sh
rustup target add wasm32-unknown-unknown --toolchain stable
```

The auto splitter can now be compiled:
```sh
cargo b --release
```

The auto splitter is then available at:
```
target/wasm32-unknown-unknown/release/silksong_autosplit_wasm.wasm
```

Make sure to look into the [API documentation](https://livesplit.org/asr/asr/) for the `asr` crate.

## Development

You can use the [debugger](https://github.com/LiveSplit/asr-debugger) while
developing the auto splitter to more easily see the log messages, statistics,
dump memory, step through the code and more.

The repository comes with preconfigured Visual Studio Code tasks. During
development it is recommended to use the `Debug Auto Splitter` launch action to
run the `asr-debugger`. You need to install the `CodeLLDB` extension to run it.

You can then use the `Build Auto Splitter (Debug)` task to manually build the
auto splitter. This will automatically hot reload the auto splitter in the
`asr-debugger`.

Alternatively you can install the [`cargo
watch`](https://github.com/watchexec/cargo-watch?tab=readme-ov-file#install)
subcommand and run the `Watch Auto Splitter` task for it to automatically build
when you save your changes.

The debugger is able to step through the code. You can set breakpoints in VSCode
and it should stop there when the breakpoint is hit. Inspecting variables may
not work all the time.

## Contributing

My approach to adding a new autosplit would look like this:
1. Search through the list of fields ([Silksong-Mono-dissector.TXT](Silksong-Mono-dissector.TXT)) to find one or more candidate fields that might correspond to what the autosplit should look for. For example on `Silk Spear (Skill)`, my candidate fields were `hasSilkSpecial` and `hasNeedleThrow`, and I wasn't sure which was the right one.
2. Test all candidate fields using a testing tool (https://github.com/AlexKnauth/asr-unity-mono-mac-testing/tree/silksong in combination with https://github.com/LiveSplit/asr-debugger can test it on all 3 OS's, not just Mac), ideally playing the game from the point right before getting to the point you want, seeing that good candidates should be `false` before, and then once you get the skill or boss or whatever, good candidates should be `true` after. Even better to test using a 2nd moniter so you can see exactly when a field goes from `false` to `true`. After I did this for `hasSilkSpecial` and `hasNeedleThrow`, I saw both go from `false` to `true` at basically the same time, so this didn't actually narrow it down, but at least confirmed they were related.
3. If multiple candidates pass step (2), ask for help. In the example of `hasSilkSpecial` and `hasNeedleThrow`, I got help from Atomic and Kazekai on the speedrun discord `#ss-tech-support` channel.
4. Make a new branch on your clone of the Github repository for the new feature you want to add. I'd recommend that you *don't* just use your master branch.
5. Add the field to the relevant `declare_pointers!` statement in `silksong_memory.rs`, add the split to the `Splits` datatype in `splits.rs`, and add the code for the split in the relevant function (either `menu_splits`, `transition_splits`, or `continuous_splits` in `splits.rs`).
6. Do not update `splits.json`, unless you are deploying a new release, in which case see below.
7. Make a Pull Request on the Github repository (https://github.com/AlexKnauth/silksong-autosplit-wasm/pulls).

## Deploying a new release

My approach to deploying a new release looks like this:
1. Review Pull Requests, and merge those that are good and ready to the master branch.
2. Update `splits.json` with the command `make examples/splits.json`. If `make` says it's up-to-date and you know it isn't, `touch src/splits.rs` before running `make` again.
3. Update `Cargo.toml` with the new version number, following [Semantic Versioning](https://semver.org/). Given `MAJOR.MINOR.PATCH`:
   - Increment `PATCH` when just releasing bug-fixes that don't add any new settings or splits.
   - Increment `MINOR` when the release includes new features, new settings, or new splits.
   - Increment `MAJOR` if there've been incompatible settings changes, but like... try to avoid those if possible.
4. Run `cargo b` in both debug and `--release` mode, and in both `--no-default-features` and default-features mode. Check that `Cargo.lock` has been updated.
5. Commit those changes, which should include `splits.json`, `Cargo.toml`, and `Cargo.lock`, with a commit name starting with `Release` and then the new version number.
6. Add a tag with the new version number on the Release commit, and push both master and the tag.
7. Check the CI to make sure all jobs pass. Sometimes there's a data race between jobs for `legacy` vs `stable`, where they both try to create their own release at the same time, so a release with only one of them is created as the other fails. When this happens, re-run the failed jobs to ensure that the release contains both `legacy` and `stable` variants.

silksong-autosplit-wasm

silksong-autosplit-wasm 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

silksong-autosplit-wasm

silksong-autosplit-wasm 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README