Fatal error: Index out of range while trying to parse non valid HTML #344
Description
This is the error described in #189
As an example this HTML will throw an error:
public func matchesAny(_ seq: ParsingStrings) -> Bool {
var buffer = [UInt8](repeating: 0, count: 4) // Max UTF-8 sequence is 4 bytes
var length = 1
buffer[0] = input[pos]
// Check if the first byte indicates a multi-byte character
if buffer[0] & 0b10000000 != 0 {
if buffer[0] & 0b11100000 == 0b11000000, pos + 1 < end {
buffer[1] = input[pos + 1]
length = 2
} else if buffer[0] & 0b11110000 == 0b11100000, pos + 2 < end {
buffer[1] = input[pos + 1]
buffer[2] = input[pos + 2]
length = 3
} else if buffer[0] & 0b11111000 == 0b11110000, pos + 3 < end {
buffer[1] = input[pos + 1]
buffer[2] = input[pos + 2]
buffer[3] = input[pos + 3]
length = 4
} else {
return false // Invalid UTF-8 sequence
}
}
let bufferSlice = buffer[..<length]
return seq.contains(bufferSlice)
}
Specifically in buffer[0] = input[pos]