Re: [vwrap] Removing first name / last name assumptions?

[Carlo Wood <[e-mail]>]

On Tue, Apr 06, 2010 at 12:32:08AM -0400, Barry Leiba wrote:
> This scheme only adds two characters to the identifier (the leading
> and trailing delimiter, and we *could* eliminate the trailing one, if
> people prefer, by defining agent-identifier as "1*(delim id-part)"...
> I like having it), and seems to solve everyone's problem.

It solves Linden Labs "problem". I mean, you are not suggesting that
this should be part of VWRAP right?

VWRAP doesn't need to specify how each login server will decode
the AgentID string. That can be different from provider to provider.


Since Meadhbh kept saying that the LL database is "dirty" so that
not ANY single delimiter will do (they have first and last names
that contain any of the 256 binary values of character), how about
encoding it like this:

[length][firstname][lastname]

where [length] is the first byte (8 bit) and represents
the length of the firstname. For example:

"\05CarloWood"

That would allow for ANY characters anywhere in both first- and
lastname.

Implementation example (in C++):

#include <iostream>

int const maxnamelen = 32;      // Must be less than 256.

int main()
{
  std::string const input_firstname = "Carlo";
  std::string const input_lastname = "Wood";

  // Encode LL name pair to agentID for VWRAP.
  std::string agentID;
  agentID = (unsigned char)input_firstname.length();    // Never 0.
  agentID += input_firstname;
  agentID += input_lastname;

  // Decode VWRAP agentID to LL name pair.
  size_t firstlen = agentID[0];
  std::string firstname = agentID.substr(1, firstlen);
  std::string lastname = agentID.substr(firstlen + 1, agentID.length() - 1 - firstlen);

  std::cout << "Firstname = \"" << firstname << "\"; lastname = \"" << lastname << "\"." << std::endl;
}

-- 
Carlo Wood <[e-mail]>