TextEncodingName

TextEncodingName = "utf-8" | "latin1" | "ascii" | "utf-16be" | "utf-32be"

Defined in: packages/core/src/buffer/text/text-encoding-name.ts:19

Text encodings encodeText/decodeText support without a platform-specific decoder.

"utf-8" and "utf-16be" are backed by a real TextDecoder instance (see decodeText’s options for their fatal/ignoreBOM support); "utf-8" is also the only encoding encodeText can use a real TextEncoder for, since TextEncoder itself only ever produces UTF-8.

"latin1"/"ascii" are one byte per UTF-16 code unit ("ascii" additionally masks the high bit, matching Node’s Buffer behaviour) — deliberately not the WHATWG Encoding Standard’s "latin1"/"ascii" labels, which that standard defines as aliases for windows-1252, not true ISO-8859-1/ASCII (see DecodeTextOptions’s doc for why that matters here). "utf-32be" is four bytes per Unicode code point, big-endian (ASN.1 UniversalString) — deliberately outside the WHATWG Encoding Standard, which excludes UTF-32 entirely.