Mihaela Mihaljevic
8d30010e0d
- README, roadmap: XSTS status quad and gates table now read the enforced 0/43/0/31 (Tests/XSTSSuiteTests.swift); test count 986/169 to 1448/258; encoding list and the remaining-work section reflect shipped reality; check-all expansion lists the validation and changelog gates - GEMINI.md, onboarding-antigravity.md: valid-schemas-rejected 1 to 0 and the pinned baselines updated to current - SECURITY.md: complete the reporting address - Parsing.InputEncoding: doc comment lists the full shipped encoding set - Schema.XSDParserUnion: guard the localName interpolation with ?? "" to match the same line and silence an optional-interpolation warning (internal dedup key, behavior preserving) - CHANGELOG: Unreleased entries for the above
91 lines
3.4 KiB
Swift
91 lines
3.4 KiB
Swift
public extension PureXML.Parsing {
|
|
/// A character encoding PureXML can decode from raw bytes. Covers the Unicode
|
|
/// transformation formats, the ISO-8859 single-byte family, the Windows code
|
|
/// pages (1250-1258), KOI8-R/U, and the CJK multi-byte encodings (Shift-JIS,
|
|
/// EUC-JP/KR/TW, GBK, GB18030, Big5, ISO-2022-JP), for libxml2 encoding parity.
|
|
enum InputEncoding: String, Equatable, Sendable {
|
|
case utf8
|
|
case utf16BigEndian
|
|
case utf16LittleEndian
|
|
case utf32BigEndian
|
|
case utf32LittleEndian
|
|
case latin1
|
|
case windows1252
|
|
case windows1254 // Turkish
|
|
case latin2 // ISO-8859-2
|
|
case latin3 // ISO-8859-3
|
|
case latin4 // ISO-8859-4
|
|
case greek // ISO-8859-7
|
|
case latin7 // ISO-8859-13
|
|
case arabic // ISO-8859-6
|
|
case hebrew // ISO-8859-8
|
|
case latin6 // ISO-8859-10
|
|
case latin8 // ISO-8859-14
|
|
case latin10 // ISO-8859-16
|
|
case thai // ISO-8859-11
|
|
case windows1250
|
|
case windows1251
|
|
case windows1253
|
|
case windows1257
|
|
case windows1255
|
|
case windows1256
|
|
case windows1258
|
|
case koi8r
|
|
case koi8u
|
|
case shiftJIS
|
|
case eucJP
|
|
case eucKR
|
|
case gbk
|
|
case gb18030
|
|
case big5
|
|
case iso2022jp
|
|
case eucTW
|
|
case latinCyrillic // ISO-8859-5
|
|
case latin5 // ISO-8859-9 (Turkish)
|
|
case latin9 // ISO-8859-15
|
|
|
|
/// Detects the encoding from the leading bytes, following the XML sniff
|
|
/// order: a byte-order mark first, then the `<?` byte pattern of the XML
|
|
/// declaration, defaulting to UTF-8. Single-byte encodings are not
|
|
/// detectable from bytes alone; they come from the declared encoding name.
|
|
public static func detect(_ bytes: [UInt8]) -> InputEncoding {
|
|
detectWithBOM(bytes).encoding
|
|
}
|
|
|
|
/// Detects the encoding and the length of any byte-order mark to strip.
|
|
static func detectWithBOM(_ bytes: [UInt8]) -> (encoding: InputEncoding, bomLength: Int) {
|
|
if bytes.starts(with: [0xEF, 0xBB, 0xBF]) {
|
|
return (.utf8, 3)
|
|
}
|
|
// UTF-32 marks are four bytes and must be checked before UTF-16, since
|
|
// the UTF-32LE mark begins with the UTF-16LE mark.
|
|
if bytes.starts(with: [0x00, 0x00, 0xFE, 0xFF]) {
|
|
return (.utf32BigEndian, 4)
|
|
}
|
|
if bytes.starts(with: [0xFF, 0xFE, 0x00, 0x00]) {
|
|
return (.utf32LittleEndian, 4)
|
|
}
|
|
if bytes.starts(with: [0xFE, 0xFF]) {
|
|
return (.utf16BigEndian, 2)
|
|
}
|
|
if bytes.starts(with: [0xFF, 0xFE]) {
|
|
return (.utf16LittleEndian, 2)
|
|
}
|
|
// No BOM: sniff the "<?" byte pattern of an XML declaration.
|
|
if bytes.starts(with: [0x00, 0x00, 0x00, 0x3C]) {
|
|
return (.utf32BigEndian, 0)
|
|
}
|
|
if bytes.starts(with: [0x3C, 0x00, 0x00, 0x00]) {
|
|
return (.utf32LittleEndian, 0)
|
|
}
|
|
if bytes.starts(with: [0x00, 0x3C, 0x00, 0x3F]) {
|
|
return (.utf16BigEndian, 0)
|
|
}
|
|
if bytes.starts(with: [0x3C, 0x00, 0x3F, 0x00]) {
|
|
return (.utf16LittleEndian, 0)
|
|
}
|
|
return (.utf8, 0)
|
|
}
|
|
}
|
|
}
|