Mailing-List: contact security-dev-help@xml.apache.org; run by ezmlm
Precedence: bulk
Reply-To: security-dev@xml.apache.org
Received-SPF: pass (asf.osuosl.org: local policy)
Message-ID: <42C92CDD.5080702@wingsofhermes.org>
Date: Mon, 04 Jul 2005 22:34:37 +1000
From: Berin Lautenbach <berin@wingsofhermes.org>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US;
 rv:1.7.3) Gecko/20040910
MIME-Version: 1.0
To: security-dev@xml.apache.org
Subject: Use of URIs rather than enums in C++ library
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit

Peoples,

I have done some work to implement an algorithm mapper for the signature 
and digest algorithms in the C++ library.  As part of this, I have 
implemented new methods that allow callers to create new signatures or 
references using URIs rather than the current enumerated type.   This is 
for two reasons

1.  The algorithm mapper makes the library more extensible - you can 
implement a new algorithm (with associated URI) and register it with the 
library.  Thus the library does not have to know about a particular 
algorithm for it to be useable.

2.  The enumerated types become a bit more hidden to callers - you don't 
need to know the algorithm enum, you just have to pass in the standard URI.

The reason for the e-mail - I'm going to check all this in later in the 
week.  I've marked the old signature and reference creation calls (using 
enums) as deprecated, as it was probably a bad idea to go down that path 
in the first place.

But does anyone feel they (the enum based creation calls) should be left 
in-perpetuity?  (They won't be removed quickly, but they will eventually 
disappear.)

Cheers,
	Berin