[Stdlib] Document StringSlice[byte=] subscript performance in docstring

[msaelices]

Summary

Add a note to StringSlice.__getitem__(byte=) docstring explaining that the UTF-8 boundary check runs in release builds, and pointing users to as_bytes() for performance-sensitive byte-scanning loops.

Assisted-by: AI

[Copilot]

Pull request overview

Updates the StringSlice.__getitem__(byte=) docstring to clarify performance implications of byte subscripting, specifically calling out the UTF-8 boundary check and recommending as_bytes() for tight byte-scanning loops.

Changes:

• Added a docstring note that __getitem__(byte=) performs a UTF-8 codepoint-boundary check in typical builds.
• Pointed users to as_bytes() / Span[Byte] for performance-sensitive byte iteration.

Comments suppressed due to low confidence (1)

mojo/stdlib/std/collections/string/string_slice.mojo:1133

• The docstring describes this as "byte-level indexing" that may return a partial/invalid UTF-8 sequence and says the return is a "String containing a single byte", but the implementation requires a UTF-8 start byte (via debug_assert[assert_mode="safe"]) and returns a StringSlice whose length is the full codepoint byte length. Please update the docstring to reflect that this returns the codepoint slice at a given byte offset (and that the offset must be a codepoint boundary).

        """Gets a single byte at the specified byte index.

        This performs byte-level indexing, not character (codepoint) indexing.
        For strings containing multi-byte UTF-8 characters, this may return a
        partial or invalid character sequence. For proper character access, use
        `codepoint_slices()` or iterate over the string directly.

        This method includes a UTF-8 codepoint boundary check that runs even
        in release builds. For performance-sensitive byte-scanning loops (e.g.
        parsers, regex engines), use `as_bytes()` to access the underlying
        `Span[Byte]` directly, which avoids this overhead.

        Parameters:
            I: A type that can be used as an index.

        Args:
            byte: The byte index (0-based). Negative indices count from the end.

        Returns:
            A new String containing a single byte at the specified position.
        """

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The new note says the UTF-8 boundary check "runs even in release builds", but the code is a debug_assert[assert_mode="safe"], which can be compiled out when assertions are disabled (e.g. assert_mode=none). Consider rewording to something like "a debug_assert[assert_mode=\"safe\"] boundary check (enabled by default, including in release builds)" to avoid overstating the guarantee.

Suggested change

	This method includes a UTF-8 codepoint boundary check that runs even
	in release builds. For performance-sensitive byte-scanning loops (e.g.
	This method includes a `debug_assert[assert_mode="safe"]` UTF-8
	codepoint boundary check (enabled by default, including in release
	builds). For performance-sensitive byte-scanning loops (e.g.

[NathanSWard]

!sync