You are viewing a plain text version of this content. The canonical link for it is here.
Posted to proton@qpid.apache.org by "Darryl L. Pierce" <dp...@redhat.com> on 2013/02/22 16:40:27 UTC

UUID size

In writing the Qpid::Proton::Data class I hit a stumbling block with
handling the UUID node type. The pn_uuid_t type is a 16-byte struct, but
what we get in Ruby is a 32 character string [1] which generates a V4
UUID.

So my question is, what sort of UUID are we trying to represent? With
the size in the pn_uuid_t type we can only hold a V1 UUID. I think we
should expand it to be a 32-bytes to hold larger modern UUIDs.

[1] http://www.ruby-doc.org/stdlib-1.9.3/libdoc/securerandom/rdoc/SecureRandom.html#method-c-uuid

-- 
Darryl L. Pierce, Sr. Software Engineer @ Red Hat, Inc.
Delivering value year after year.
Red Hat ranks #1 in value among software vendors.
http://www.redhat.com/promo/vendor/


Re: UUID size

Posted by "Darryl L. Pierce" <dp...@redhat.com>.
On Fri, Feb 22, 2013 at 11:29:44AM -0500, Ken Giusti wrote:
> Hi Darryl,
> 
> Proton represents the UUID as binary, hence the 16 byte array.  For string representations, such as Ruby's, that would have to be converted to ascii/utf8 hex, and would require the 32 bytes for proper representation.
> 
> In python, we have to manually convert the uuid in pn_data_t format into the native python uuid object type - see bindings/python/proton.py.  Sounds like you'll have something similar in ruby.
> 
> Aside, I think the V# type just indicates the method used for generating the 16  byte uuid.  V4 uses a random number generator, for example.  I think the sizes are all 16 byte.

Thanks, Ken. I'll look at the Python bindings.

-- 
Darryl L. Pierce, Sr. Software Engineer @ Red Hat, Inc.
Delivering value year after year.
Red Hat ranks #1 in value among software vendors.
http://www.redhat.com/promo/vendor/


Re: UUID size

Posted by Ted Ross <tr...@redhat.com>.
Also,  the AMQP specification defines uuid as 16-bytes.

-Ted

On 02/22/2013 11:29 AM, Ken Giusti wrote:
> Hi Darryl,
>
> Proton represents the UUID as binary, hence the 16 byte array.  For string representations, such as Ruby's, that would have to be converted to ascii/utf8 hex, and would require the 32 bytes for proper representation.
>
> In python, we have to manually convert the uuid in pn_data_t format into the native python uuid object type - see bindings/python/proton.py.  Sounds like you'll have something similar in ruby.
>
> Aside, I think the V# type just indicates the method used for generating the 16  byte uuid.  V4 uses a random number generator, for example.  I think the sizes are all 16 byte.
>
> -K
>
> ----- Original Message -----
>> In writing the Qpid::Proton::Data class I hit a stumbling block with
>> handling the UUID node type. The pn_uuid_t type is a 16-byte struct,
>> but
>> what we get in Ruby is a 32 character string [1] which generates a V4
>> UUID.
>>
>> So my question is, what sort of UUID are we trying to represent? With
>> the size in the pn_uuid_t type we can only hold a V1 UUID. I think we
>> should expand it to be a 32-bytes to hold larger modern UUIDs.
>>
>> [1]
>> http://www.ruby-doc.org/stdlib-1.9.3/libdoc/securerandom/rdoc/SecureRandom.html#method-c-uuid
>>
>> --
>> Darryl L. Pierce, Sr. Software Engineer @ Red Hat, Inc.
>> Delivering value year after year.
>> Red Hat ranks #1 in value among software vendors.
>> http://www.redhat.com/promo/vendor/
>>
>>


Re: UUID size

Posted by Ken Giusti <kg...@redhat.com>.
Hi Darryl,

Proton represents the UUID as binary, hence the 16 byte array.  For string representations, such as Ruby's, that would have to be converted to ascii/utf8 hex, and would require the 32 bytes for proper representation.

In python, we have to manually convert the uuid in pn_data_t format into the native python uuid object type - see bindings/python/proton.py.  Sounds like you'll have something similar in ruby.

Aside, I think the V# type just indicates the method used for generating the 16  byte uuid.  V4 uses a random number generator, for example.  I think the sizes are all 16 byte.

-K

----- Original Message -----
> In writing the Qpid::Proton::Data class I hit a stumbling block with
> handling the UUID node type. The pn_uuid_t type is a 16-byte struct,
> but
> what we get in Ruby is a 32 character string [1] which generates a V4
> UUID.
> 
> So my question is, what sort of UUID are we trying to represent? With
> the size in the pn_uuid_t type we can only hold a V1 UUID. I think we
> should expand it to be a 32-bytes to hold larger modern UUIDs.
> 
> [1]
> http://www.ruby-doc.org/stdlib-1.9.3/libdoc/securerandom/rdoc/SecureRandom.html#method-c-uuid
> 
> --
> Darryl L. Pierce, Sr. Software Engineer @ Red Hat, Inc.
> Delivering value year after year.
> Red Hat ranks #1 in value among software vendors.
> http://www.redhat.com/promo/vendor/
> 
>