Parsing the output of --list-keys

[Werner Koch]

On Mon, 17 Jan 2000, Jonas Steverud wrote:

> <tnelson@conceptech.com>" is not hard since the number of characters
> from the left marging is 31 (i.e. (forward-char 31)). The problem is
> to get "Tony Nelson", "(optional comment)" and
> "tnelson@conceptech.com".

What you should do here is RFC822 address parsing.  Doing it with r.e.
is not trivial.  However, if this user ID is GnuPG generated it is
easy to get all three parts:

 1. part:  Collect everything up to (but excluding) the first
           opening parenthesis or the first left angle bracket.
 2. part:  Only if you stopped at a parenthesis, collect everthing up
	   to the closing parenthesis.  Although it is not yet
           allowed, you might want to keep a counter of opening
           closing parenthesis.
 3. part:  Only if you have a left angle bracket, collect everything
	   up to the right angle bracket.

It should be easy to construct a sed,awk,perl r.e. for this.

> I have found a way of doing this with regexps so the problem is
> solved, but if --with-colons had put colons between the wanted parts

No that is not possible.


-- 
Werner Koch at guug.de           www.gnupg.org           keyid 621CC013
  
     Boycott Amazon!  -  http://www.gnu.org/philosophy/amazon.html