Files
Odin/core/encoding/base32/base32.odin
T
Zoltán Kéri b9338777e3 encoding/base32: Fix buffer allocation and bounds checking
Fix buffer allocation size calculation and add proper bounds checking to
ensure output buffer has sufficient space. This fixes crashes that could
occur with inputs like "AA" and other edge cases where the output buffer
was too small.

Remove #no_bounds_check as proper bounds checking is necessary for safe
error handling. The small performance trade-off is worth the improved
robustness.
2024-12-24 02:20:32 +01:00

189 lines
5.4 KiB
Odin

package encoding_base32
// @note(zh): Encoding utility for Base32
// A secondary param can be used to supply a custom alphabet to
// @link(encode) and a matching decoding table to @link(decode).
// If none is supplied it just uses the standard Base32 alphabet.
// Incase your specific version does not use padding, you may
// truncate it from the encoded output.
// Error represents errors that can occur during base32 decoding operations.
// See RFC 4648 sections 3.2, 4 and 6.
Error :: enum {
None,
Invalid_Character, // Input contains characters outside of base32 alphabet (A-Z, 2-7)
Invalid_Length, // Input length is not valid for base32 (must be a multiple of 8 with proper padding)
Malformed_Input, // Input has improper structure (wrong padding position or incomplete groups)
}
ENC_TABLE := [32]byte {
'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H',
'I', 'J', 'K', 'L', 'M', 'N', 'O', 'P',
'Q', 'R', 'S', 'T', 'U', 'V', 'W', 'X',
'Y', 'Z', '2', '3', '4', '5', '6', '7',
}
PADDING :: '='
DEC_TABLE := [?]u8 {
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 26, 27, 28, 29, 30, 31, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 0, 0, 0, 0, 0,
0, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
}
encode :: proc(data: []byte, ENC_TBL := ENC_TABLE, allocator := context.allocator) -> string {
out_length := (len(data) + 4) / 5 * 8
out := make([]byte, out_length)
_encode(out, data)
return string(out)
}
@private
_encode :: proc(out, data: []byte, ENC_TBL := ENC_TABLE, allocator := context.allocator) {
out := out
data := data
for len(data) > 0 {
carry: byte
switch len(data) {
case:
out[7] = ENC_TABLE[data[4] & 0x1f]
carry = data[4] >> 5
fallthrough
case 4:
out[6] = ENC_TABLE[carry | (data[3] << 3) & 0x1f]
out[5] = ENC_TABLE[(data[3] >> 2) & 0x1f]
carry = data[3] >> 7
fallthrough
case 3:
out[4] = ENC_TABLE[carry | (data[2] << 1) & 0x1f]
carry = (data[2] >> 4) & 0x1f
fallthrough
case 2:
out[3] = ENC_TABLE[carry | (data[1] << 4) & 0x1f]
out[2] = ENC_TABLE[(data[1] >> 1) & 0x1f]
carry = (data[1] >> 6) & 0x1f
fallthrough
case 1:
out[1] = ENC_TABLE[carry | (data[0] << 2) & 0x1f]
out[0] = ENC_TABLE[data[0] >> 3]
}
if len(data) < 5 {
out[7] = byte(PADDING)
if len(data) < 4 {
out[6] = byte(PADDING)
out[5] = byte(PADDING)
if len(data) < 3 {
out[4] = byte(PADDING)
if len(data) < 2 {
out[3] = byte(PADDING)
out[2] = byte(PADDING)
}
}
}
break
}
data = data[5:]
out = out[8:]
}
}
decode :: proc(data: string, DEC_TBL := DEC_TABLE, allocator := context.allocator) -> ([]byte, Error) {
if len(data) == 0 {
return nil, .None
}
// Calculate maximum possible output size and allocate buffer
out_len := (len(data) * 5 + 7) / 8 // Ceiling division to ensure enough space
out := make([]byte, out_len, allocator)
outi := 0
data := data
end := false
for len(data) > 0 && !end {
dbuf : [8]byte
dlen := 8
for j := 0; j < 8; {
if len(data) == 0 {
dlen, end = j, true
break
}
input := data[0]
data = data[1:]
if input == byte(PADDING) && j >= 2 && len(data) < 8 {
if len(data) + j < 8 - 1 {
return nil, .Malformed_Input
}
for k := 0; k < 8-1-j; k += 1 {
if len(data) < k || data[k] != byte(PADDING) {
return nil, .Malformed_Input
}
}
dlen, end = j, true
if dlen == 1 || dlen == 3 || dlen == 6 {
return nil, .Invalid_Length
}
break
}
decoded := DEC_TBL[input]
if decoded == 0 && input != byte(ENC_TABLE[0]) {
return nil, .Invalid_Character
}
dbuf[j] = decoded
j += 1
}
// Ensure we have enough space in output buffer
needed := 5 // Each full 8-char block produces 5 bytes
if outi + needed > len(out) {
return nil, .Invalid_Length
}
// Process complete input blocks
switch dlen {
case 8:
if len(dbuf) < 8 { return nil, .Invalid_Length }
out[outi + 4] = dbuf[6] << 5 | dbuf[7]
fallthrough
case 7:
if len(dbuf) < 7 { return nil, .Invalid_Length }
out[outi + 3] = dbuf[4] << 7 | dbuf[5] << 2 | dbuf[6] >> 3
fallthrough
case 5:
if len(dbuf) < 5 { return nil, .Invalid_Length }
out[outi + 2] = dbuf[3] << 4 | dbuf[4] >> 1
fallthrough
case 4:
if len(dbuf) < 4 { return nil, .Invalid_Length }
out[outi + 1] = dbuf[1] << 6 | dbuf[2] << 1 | dbuf[3] >> 4
fallthrough
case 2:
if len(dbuf) < 2 { return nil, .Invalid_Length }
out[outi + 0] = dbuf[0] << 3 | dbuf[1] >> 2
}
outi += 5
}
// Trim output buffer to actual size
if outi < len(out) {
out = out[:outi]
}
return out, .None
}