Re: Aw: Re: gpgme replacement



On Thu, 2003-03-27 at 04:40, albrecht.dress@arcor.de wrote:
> > I disagree your point of view about GPGME :
> [snipped Laurent's reply]
> > I don't know Albrecht opinion but for me, I prefer to continue to use GPGME
> > directly.
> 
> I can just tell you that I agree fully with Laurent, and I couldn't say it better!!! I also think that gpgme is nice and (after digging through the docs ;-)) easy to use. When the mt issues are resolved (yes, I found a hang on an old SuSE 7.2 system - already reported it to the gpgme guys!), it seems to be perfect for me.
> 
> There might be, however, some **ui** related things where a seahorse if could make sense:
> 
> * reading the passphrase. This is probabely the most critical point in the whole process, as some nasty sysop might just kill your app and try to get it from the core or the swap space. I try to protect it by first overwriting it with random data before I free it, but some daemon which is not swapped (and needs root privileges for that) would be nice.

I believe that gpg-2 will be able to do this, but that won't be
available for a while.  Right now there is a gpg-agent in the newpg
package that aims to provide this feature (AFAIK).

> * (not so important) let seahorse select the key if there is more than one for a sender/recipient, and let seahorse try to get a missing key from a keyserver (but afaik it does not support the keyserver yet?)

So if seahorse was installed and exported some kind of API for getting
keys from a keyserver, then maybe balsa would use it?
Maybe a better option would be to help me pester the gpgme folks to add
keyserver support so I don't have to ;)

One ui bug I noticed is that when encryption fails due to a bad
passphrase, two dialogs popup instead of one. Bugzilla?
-- 
Jacob Perkins <jap1@users.sourceforge.net>
seahorse



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]