On 13/3/06 5:57, Diffenderfer, Randy <[EMAIL PROTECTED]> wrote:
> Folks,
>
> Shouldn't have hit send so quickly. A couple more things of interest in
> comparing the dump routines of Convert::ASN1 and Convert::BER. I
> realize these are for *debug only*, per the source, but ...
>
> In Convert::BER->dump(), I think that most of the addresses of the
> contents of "complicated" dumped objects are "off", pointing at the
> length octet(s).
>
> In both routines, however, the handling of this INTEGER is suspect:
>
> 0085 02 16: INTEGER = 1766191625
> 0097 : }
> 0097 30 13: SEQUENCE {
> 0099 06 9: OBJECT ID = 1.2.840.113549.1.1.1
> 00A4 05 0: NULL
> 00A6 : }
>
> Where the raw octets were:
>
> 0080 6E 67 20 43 41 02 10 69 45 F2 09 14 55 10 D8 7F ng CA..iE...U...
> 0090 2C A2 97 D0 DD 65 CF 30 0D 06 09 2A 86 48 86 F7 ,....e.0...*.H..
>
> Here is the "openssl asn1parse" view of the same area (offsets in
> decimal...):
>
> 133:d=6 hl=2 l= 16 prim: INTEGER
> :6945F209145510D87F2CA297D0DD65CF
> 151:d=5 hl=2 l= 13 cons: SEQUENCE
> 153:d=6 hl=2 l= 9 prim: OBJECT :rsaEncryption
Convert::ASN's got a bignum mode which will allow things like stupid
Verisign serial numbers to be more easily held and (I guess) printed. (I'm
away from the docs to check this.)
Cheers,
Chris
