ASN1 syntax error

[mateipopp]

Hello,
I'm trying to decode a ANS1 encoded with BER, I edited your example and i get some output, but after some line i got a error.
This is the output:
[C] 0x15
[C] 0x0: 0x13
[C] BIT STRING: 0x01803809006312F5
[C] OCTET STRING
[C] 0x0: 0x01000002

This is the error:
File "test.py", line 133, in pretty_print
pretty_print(input_stream, output_stream, indent + 2)
File "test.py", line 126, in pretty_print
tag, value = input_stream.read()
File "Documents\ANS1\asn1.py", line 421, in read
value = self._read_value(tag.nr, length)
File "Documents\ANS1\asn1.py", line 516, in _read_value
value = self._decode_null(bytes_data)
File "Documents\ANS1\asn1.py", line 598, in _decode_null
raise Error('ASN1 syntax error')
asn1.Error: ASN1 syntax error

I have tried to use a online decoder. https://lapo.it/asn1js/
and the output is:

[21] (24 elem)
[0] (1 byte) 13
[3] (8 byte) 01803809006312F5
[4] (1 elem)
[0] (4 byte) 01000002
[5] (5 byte) 0087B31F73
[6] (1 elem)
[0] (4 byte) 01000001
[7] test.test.se
[8] (2 byte) 0121
Offset: 53
Length: 2+2
Value:
(2 byte)
0121
[9] (1 elem)
[0] (1 elem)
[0] (4 byte) 0AD88204
[11] (1 byte) 01
[12] (1 elem)
SEQUENCE (6 elem)
[2] (12 byte) 0223621F9396405874631040
[3] (1 byte) 00
[4] (1 byte) 00
[5] (1 byte) 02
[6] (9 byte) 1709150029542B0000
[8] (8 byte) 0022F210006D188F
[13] (9 byte) 1709142329542B0000
[14] (2 byte) 0E10
[15] (1 byte) 00
[17] (1 byte) 01
[18] DCP2GGSN1
[21] (1 byte) 00
[22] (9 byte) 9123770050605303F0
[23] (2 byte) 0000
[24] (1 byte) 00
[27] (3 byte) 22F210
[29] (8 byte) 5393297058824310
[30] (1 byte) 02
[31] (2 byte) 8001
[32] (8 byte) 0022F210006D188F

The ANS1 content i'm trying to decode is:
tYHKgAETgwgBgDgJAGMS9aQGgAQBAAAChQUAh7Mfc6YGgAQBAAABhwx0ZXN0LnRlc3Quc2WIAgEhqQig
BoAECtiCBIsBAawuMCyCDAIjYh+TlkBYdGMQQIMBAIQBAIUBAoYJFwkVAClUKwAAiAgAIvIQAG0Yj40J
FwkUIylUKwAAjgIOEI8BAJEBAZIJRENQMkdHU04xlQEAlgmRI3cAUGBTA/CXAgAAmAEAmwMi8hCdCFOT
KXBYgkMQngECnx8CgAGfIAgAIvIQAG0Yjw==

thanks.

[keshavwork]

i am also facing same issue can you please update the code.

[andrivet]

@keshavwork i do not have the time. PR is welcome.

[karnh]

I found a couple of issues which I've fixed in my local repo but I don't know how to create PR request. I think in above case the problem is the package is trying to decode context or application class primitive tagid as Universal primitive tagid.

def _read_value(self, nr, cls, length):  # type: (int, int) -> any
    """Read a value from the input."""
    
    bytes_data = self._read_bytes(length)
    
    # if tag class is not universal
    if cls:
        try:
            value = self._decode_printable_string(bytes_data)
        except:
            value = self._decode_integer(bytes_data)

        return value

I'm trying to decode as string first followed by integer but I think better fix will be either return raw bytes so user can decode it or let user provide type in input_stream.read() call.

Another issue I fixed is that the parser fails to understand boundary of indefinite length tags and enter() and leave() method of parser do not work.

I'll see if I can create a PR and submit it.

thanks.

[andrivet]

This is fixed in version 1.4.0. Non-Universal data are decoded as-is (i.e. as binary).