src/ContentHash.bend source

src/ContentHash.bend on the hub · documented module

import Baseimport mylsm-lsm-store@0.3.2.0/src/hub_sha/sha256.bend as ProvenShaimport bend-codec-lib@0.2.0.0/utf8.bend as Utf8Libimport bend-codec-lib@0.2.0.0/bytes.bend as ByteLib# SHA-256 hex of UTF-8 text. ProvenSha/Utf8 cores are trusted upstream# (FIPS 180-4 machine-checked; closed roundtrips proved); this module owns# only the byte-list plumbing between them, covered by the differential# check vs Python hashlib at the 55/56/64-byte padding boundaries.# Unwraps the byte box into the word list the hash core consumes.def bytes_data(byte_box: ByteLib.Bytes) -> List<&2, U32>:  match byte_box:    case ByteLib.Bytes{raw_data}:      raw_data# Hashes valid encodings; unencodable input yields a fixed sentinel string.def digest_from_encode_result(encode_result: Result<&2, &2, Utf8Lib.Utf8.Error, ByteLib.Bytes>) -> String:  match encode_result:    case Fail{encode_error}:      "INVALID_UTF8_INPUT"    case Done{byte_box}:      ProvenSha.hex(ProvenSha.sha256(bytes_data(byte_box)))# "INVALID_UTF8_INPUT" on unencodable text; never fails on valid input.def compute_sha256_hex(input_text: String) -> String:  digest_from_encode_result(Utf8Lib.Utf8.encode(input_text))