Rollup merge of #85275 - CDirkx:memchr, r=m-ou-se

Move `std::memchr` to `sys_common`

`std::memchr` is a thin abstraction over the different `memchr` implementations in `sys`, along with documentation and tests. The module is only used internally by `std`, nothing is exported externally. Code like this is exactly what the `sys_common` module is for, so this PR moves it there.

3 files changed, 138 insertions, 0 deletions

diff --git a/library/std/src/sys_common/memchr.rs b/library/std/src/sys_common/memchr.rs
new file mode 100644
index 00000000000..b219e878912
--- /dev/null
+++ b/library/std/src/sys_common/memchr.rs
@@ -0,0 +1,51 @@
+// Original implementation taken from rust-memchr.
+// Copyright 2015 Andrew Gallant, bluss and Nicolas Koch
+
+use crate::sys::memchr as sys;
+
+#[cfg(test)]
+mod tests;
+
+/// A safe interface to `memchr`.
+///
+/// Returns the index corresponding to the first occurrence of `needle` in
+/// `haystack`, or `None` if one is not found.
+///
+/// memchr reduces to super-optimized machine code at around an order of
+/// magnitude faster than `haystack.iter().position(|&b| b == needle)`.
+/// (See benchmarks.)
+///
+/// # Examples
+///
+/// This shows how to find the first position of a byte in a byte string.
+///
+/// ```ignore (cannot-doctest-private-modules)
+/// use memchr::memchr;
+///
+/// let haystack = b"the quick brown fox";
+/// assert_eq!(memchr(b'k', haystack), Some(8));
+/// ```
+#[inline]
+pub fn memchr(needle: u8, haystack: &[u8]) -> Option<usize> {
+    sys::memchr(needle, haystack)
+}
+
+/// A safe interface to `memrchr`.
+///
+/// Returns the index corresponding to the last occurrence of `needle` in
+/// `haystack`, or `None` if one is not found.
+///
+/// # Examples
+///
+/// This shows how to find the last position of a byte in a byte string.
+///
+/// ```ignore (cannot-doctest-private-modules)
+/// use memchr::memrchr;
+///
+/// let haystack = b"the quick brown fox";
+/// assert_eq!(memrchr(b'o', haystack), Some(17));
+/// ```
+#[inline]
+pub fn memrchr(needle: u8, haystack: &[u8]) -> Option<usize> {
+    sys::memrchr(needle, haystack)
+}
diff --git a/library/std/src/sys_common/memchr/tests.rs b/library/std/src/sys_common/memchr/tests.rs
new file mode 100644
index 00000000000..557d749c7f6
--- /dev/null
+++ b/library/std/src/sys_common/memchr/tests.rs
@@ -0,0 +1,86 @@
+// Original implementation taken from rust-memchr.
+// Copyright 2015 Andrew Gallant, bluss and Nicolas Koch
+
+// test the implementations for the current platform
+use super::{memchr, memrchr};
+
+#[test]
+fn matches_one() {
+    assert_eq!(Some(0), memchr(b'a', b"a"));
+}
+
+#[test]
+fn matches_begin() {
+    assert_eq!(Some(0), memchr(b'a', b"aaaa"));
+}
+
+#[test]
+fn matches_end() {
+    assert_eq!(Some(4), memchr(b'z', b"aaaaz"));
+}
+
+#[test]
+fn matches_nul() {
+    assert_eq!(Some(4), memchr(b'\x00', b"aaaa\x00"));
+}
+
+#[test]
+fn matches_past_nul() {
+    assert_eq!(Some(5), memchr(b'z', b"aaaa\x00z"));
+}
+
+#[test]
+fn no_match_empty() {
+    assert_eq!(None, memchr(b'a', b""));
+}
+
+#[test]
+fn no_match() {
+    assert_eq!(None, memchr(b'a', b"xyz"));
+}
+
+#[test]
+fn matches_one_reversed() {
+    assert_eq!(Some(0), memrchr(b'a', b"a"));
+}
+
+#[test]
+fn matches_begin_reversed() {
+    assert_eq!(Some(3), memrchr(b'a', b"aaaa"));
+}
+
+#[test]
+fn matches_end_reversed() {
+    assert_eq!(Some(0), memrchr(b'z', b"zaaaa"));
+}
+
+#[test]
+fn matches_nul_reversed() {
+    assert_eq!(Some(4), memrchr(b'\x00', b"aaaa\x00"));
+}
+
+#[test]
+fn matches_past_nul_reversed() {
+    assert_eq!(Some(0), memrchr(b'z', b"z\x00aaaa"));
+}
+
+#[test]
+fn no_match_empty_reversed() {
+    assert_eq!(None, memrchr(b'a', b""));
+}
+
+#[test]
+fn no_match_reversed() {
+    assert_eq!(None, memrchr(b'a', b"xyz"));
+}
+
+#[test]
+fn each_alignment() {
+    let mut data = [1u8; 64];
+    let needle = 2;
+    let pos = 40;
+    data[pos] = needle;
+    for start in 0..16 {
+        assert_eq!(Some(pos - start), memchr(needle, &data[start..]));
+    }
+}
diff --git a/library/std/src/sys_common/mod.rs b/library/std/src/sys_common/mod.rs
index 4ef0e72adf0..1a9caa22c92 100644
--- a/library/std/src/sys_common/mod.rs
+++ b/library/std/src/sys_common/mod.rs
@@ -25,6 +25,7 @@ pub mod bytestring;
 pub mod condvar;
 pub mod fs;
 pub mod io;
+pub mod memchr;
 pub mod mutex;
 // `doc` is required because `sys/mod.rs` imports `unix/ext/mod.rs` on Windows
 // when generating documentation.