core

array

A growable, copy-on-write sequence with value semantics.

string

String is a UTF-8 byte buffer. Byte-level access lives behind the explicit utf8() view; byte_count is the buffer length in BYTES, not user-perceived characters. Equality is byte equality: canonically equivalent strings in different normal forms (NFC "é" vs NFD "é") compare unequal.

string_builder

Unique growable UTF-8 construction. Appending is amortized O(total bytes), and finish transfers the allocation into an immutable String without copying.

text

Deep, grapheme-safe text algorithms shared by String and Substring. Character-shaped methods only mint and consume opaque StringIndex values; byte offsets remain confined to UTF8View and compatibility search.

unicode

UTF-8 and Unicode segmentation support for String's Character layer.

unicode_data

GENERATED by genunicode from UCD 17.0.0 - do not edit. Grapheme-cluster-break categories (UAX #29), one flat partition merging GraphemeClusterBreak, ExtendedPictographic, and InCB. The table is a sorted boundary list: entries of four base-128 septets encoding startcodepoint * 32 + category. See src/bin/genunicode.rs for the encoding and regeneration story.