memchr======This library provides heavily optimized routines for string search primitives.[![Build status](https://github.com/BurntSushi/memchr/workflows/ci/badge.svg)](https://github.com/BurntSushi/memchr/actions)[![Crates.io](https://img.shields.io/crates/v/memchr.svg)](https://crates.io/crates/memchr)Dual-licensed under MIT or the [UNLICENSE](https://unlicense.org/).### Documentation[https://docs.rs/memchr](https://docs.rs/memchr)### Overview* The top-level module provides routines for searching for 1, 2 or 3 bytes  in the forward or reverse direction. When searching for more than one byte,  positions are considered a match if the byte at that position matches any  of the bytes.* The `memmem` sub-module provides forward and reverse substring search  routines.In all such cases, routines operate on `&[u8]` without regard to encoding. Thisis exactly what you want when searching either UTF-8 or arbitrary bytes.### Compiling without the standard librarymemchr links to the standard library by default, but you can disable the`std` feature if you want to use it in a `#![no_std]` crate:```toml[dependencies]memchr = { version = "2", default-features = false }```On `x86_64` platforms, when the `std` feature is disabled, the SSE2 acceleratedimplementations will be used. When `std` is enabled, AVX2 acceleratedimplementations will be used if the CPU is determined to support it at runtime.SIMD accelerated routines are also available on the `wasm32` and `aarch64`targets. The `std` feature is not required to use them.When a SIMD version is not available, then this crate falls back to[SWAR](https://en.wikipedia.org/wiki/SWAR) techniques.### Minimum Rust version policyThis crate's minimum supported `rustc` version is `1.61.0`.The current policy is that the minimum Rust version required to use this cratecan be increased in minor version updates. For example, if `crate 1.0` requiresRust 1.20.0, then `crate 1.0.z` for all values of `z` will also require Rust1.20.0 or newer. However, `crate 1.y` for `y > 0` may require a newer minimumversion of Rust.In general, this crate will be conservative with respect to the minimumsupported version of Rust.### Testing strategyGiven the complexity of the code in this crate, along with the pervasive useof `unsafe`, this crate has an extensive testing strategy. It combines multipleapproaches:* Hand-written tests.* Exhaustive-style testing meant to exercise all possible branching and offset  calculations.* Property based testing through [`quickcheck`](https://github.com/BurntSushi/quickcheck).* Fuzz testing through [`cargo fuzz`](https://github.com/rust-fuzz/cargo-fuzz).* A huge suite of benchmarks that are also run as tests. Benchmarks always  confirm that the expected result occurs.Improvements to the testing infrastructure are very welcome.### Algorithms usedAt time of writing, this crate's implementation of substring search actuallyhas a few different algorithms to choose from depending on the situation.* For very small haystacks,  [Rabin-Karp](https://en.wikipedia.org/wiki/Rabin%E2%80%93Karp_algorithm)  is used to reduce latency. Rabin-Karp has very small overhead and can often  complete before other searchers have even been constructed.* For small needles, a variant of the  ["Generic SIMD"](http://0x80.pl/articles/simd-strfind.html#algorithm-1-generic-simd)  algorithm is used. Instead of using the first and last bytes, a heuristic is  used to select bytes based on a background distribution of byte frequencies.* In all other cases,  [Two-Way](https://en.wikipedia.org/wiki/Two-way_string-matching_algorithm)  is used. If possible, a prefilter based on the "Generic SIMD" algorithm  linked above is used to find candidates quickly. A dynamic heuristic is used  to detect if the prefilter is ineffective, and if so, disables it.### Why is the standard library's substring search so much slower?We'll start by establishing what the difference in performance actuallyis. There are two relevant benchmark classes to consider: `prebuilt` and`oneshot`. The `prebuilt` benchmarks are designed to measure---to the extentpossible---search time only. That is, the benchmark first starts by building asearcher and then only tracking the time for _using_ the searcher:```$ rebar rank benchmarks/record/x86_64/2023-08-26.csv --intersection -e memchr/memmem/prebuilt -e std/memmem/prebuiltEngine                       Version                   Geometric mean of speed ratios  Benchmark count------                       -------                   ------------------------------  ---------------rust/memchr/memmem/prebuilt  2.5.0                     1.03                            53rust/std/memmem/prebuilt     1.73.0-nightly 180dffba1  6.50                            53```Conversely, the `oneshot` benchmark class measures the time it takes to bothbuild the searcher _and_ use it:```$ rebar rank benchmarks/record/x86_64/2023-08-26.csv --intersection -e memchr/memmem/oneshot -e std/memmem/oneshotEngine                      Version                   Geometric mean of speed ratios  Benchmark count------                      -------                   ------------------------------  ---------------rust/memchr/memmem/oneshot  2.5.0                     1.04                            53rust/std/memmem/oneshot     1.73.0-nightly 180dffba1  5.26                            53```**NOTE:** Replace `rebar rank` with `rebar cmp` in the above commands toexplore the specific benchmarks and their differences.So in both cases, this crate is quite a bit faster over a broad sampling ofbenchmarks regardless of whether you measure only search time or search timeplus construction time. The difference is a little smaller when you includeconstruction time in your measurements.These two different types of benchmark classes make for a nice segue intoone reason why the standard library's substring search can be slower: APIdesign. In the standard library, the only APIs available to you requireone to re-construct the searcher for every search. While you can benefitfrom building a searcher once and iterating over all matches in a singlestring, you cannot reuse that searcher to search other strings. This mightcome up when, for example, searching a file one line at a time. You'll needto re-build the searcher for every line searched, and this can [reallymatter][burntsushi-bstr-blog].**NOTE:** The `prebuilt` benchmark for the standard library can't actuallyavoid measuring searcher construction at some level, because there is no APIfor it. Instead, the benchmark consists of building the searcher once and thenfinding all matches in a single string via an iterator. This tends toapproximate a benchmark where searcher construction isn't measured, but itisn't perfect. While this means the comparison is not strictlyapples-to-apples, it does reflect what is maximally possible with the standardlibrary, and thus reflects the best that one could do in a real world scenario.While there is more to the story than just API design here, it's important topoint out that even if the standard library's substring search were a preciseclone of this crate internally, it would still be at a disadvantage in someworkloads because of its API. (The same also applies to C's standard library`memmem` function. There is no way to amortize construction of the searcher.You need to pay for it on every call.)The other reason for the difference in performance is thatthe standard library has trouble using SIMD. In particular, substring searchis implemented in the `core` library, where platform specific code generallycan't exist. That's an issue because in order to utilize SIMD beyond SSE2while maintaining portable binaries, one needs to use [dynamic CPU featuredetection][dynamic-cpu], and that in turn requires platform specific code.While there is [an RFC for enabling target feature detection in`core`][core-feature], it doesn't yet exist.The bottom line here is that `core`'s substring search implementation islimited to making use of SSE2, but not AVX.Still though, this crate does accelerate substring search even when only SSE2is available. The standard library could therefore adopt the techniques in thiscrate just for SSE2. The reason why that hasn't happened yet isn't totallyclear to me. It likely needs a champion to push it through. The standardlibrary tends to be more conservative in these things. With that said, thestandard library does use some [SSE2 acceleration on `x86-64`][std-sse2] addedin [this PR][std-sse2-pr]. However, at the time of writing, it is only usedfor short needles and doesn't use the frequency based heuristics found in thiscrate.**NOTE:** Another thing worth mentioning is that the standard library'ssubstring search routine requires that both the needle and haystack have type`&str`. Unless you can assume that your data is valid UTF-8, building a `&str`will come with the overhead of UTF-8 validation. This may in turn result inoverall slower searching depending on your workload. In contrast, the `memchr`crate permits both the needle and the haystack to have type `&[u8]`, where`&[u8]` can be created from a `&str` with zero cost. Therefore, the substringsearch in this crate is strictly more flexible than what the standard libraryprovides.[burntsushi-bstr-blog]: https://blog.burntsushi.net/bstr/#motivation-based-on-performance[dynamic-cpu]: https://doc.rust-lang.org/std/arch/index.html#dynamic-cpu-feature-detection[core-feature]: https://github.com/rust-lang/rfcs/pull/3469[std-sse2]: https://github.com/rust-lang/rust/blob/bf9229a2e366b4c311f059014a4aa08af16de5d8/library/core/src/str/pattern.rs#L1719-L1857[std-sse2-pr]: https://github.com/rust-lang/rust/pull/103779