#web #wasm #dom

macro sledgehammer_bindgen

Fast batched js bindings

4 releases

0.1.3 Dec 10, 2022
0.1.2 Dec 8, 2022
0.1.1 Dec 6, 2022
0.1.0 Dec 3, 2022

#109 in #dom

Download history 38/week @ 2022-11-30 192/week @ 2022-12-07 181/week @ 2022-12-14 148/week @ 2022-12-21 193/week @ 2022-12-28 251/week @ 2023-01-04 501/week @ 2023-01-11 246/week @ 2023-01-18 278/week @ 2023-01-25 317/week @ 2023-02-01 329/week @ 2023-02-08 451/week @ 2023-02-15 458/week @ 2023-02-22 518/week @ 2023-03-01 499/week @ 2023-03-08 378/week @ 2023-03-15

1,911 downloads per month
Used in 6 crates (via dioxus-interpreter-js)

MIT license

64KB
1.5K SLoC

sledgehammer bindgen

What is Sledgehammer Bindgen?

Sledgehammer bindgen provides faster rust batched bindings for js code.

How does this compare to wasm-bindgen:

  • wasm-bindgen is a lot more general it allows returning values and passing around a lot more different types of values. For most users wasm-bindgen is a beter choice. Sledgehammer is specifically designed for web frameworks that want low-level, fast access to the dom.

  • You can use sledgehammer bindgen with wasm-bindgen. See the docs and examples for more information.

Why is it fast?

String decoding

  • Decoding strings are expensive to decode, but the cost doesn't change much with the size of the string. Wasm-bindgen calls TextDecoder.decode for every string. Sledgehammer only calls TextEncoder.decode once per batch.

  • If the string is small, it is faster to decode the string in javascript to avoid the constant overhead of TextDecoder.decode

  • See this benchmark: https://jsbench.me/4vl97c05lb/5

String Caching

  • You can cache strings in javascript to avoid decoding the same string multiple times.
  • If the string is static sledgehammer will hash by pointer instead of by value.

Byte encoded operations

  • In sledgehammer every operation is encoded as a sequence of bytes packed into an array. Every operation takes 1 byte plus whatever data is required for it.

  • Each operation is encoded in a batch of four as a u32. Getting a number from an array buffer has a high constant cost, but getting a u32 instead of a u8 is not more expensive. Sledgehammer bindgen reads the u32 and then splits it into the 4 individual bytes. It will shuffle and pack the bytes into as few buckets as possible and try to inline reads into js.

  • See this benchmark: https://jsbench.me/csl9lfauwi/2

Dependencies

~0.7–1MB
~26K SLoC