[PATCH] Add signal handlers for controlling password entry visibility.

Chris Lumens clumens at redhat.com
Wed Apr 10 18:25:11 UTC 2013


> But if we're showing the passphrase while you're setting it, do we still
> need to make the user type it twice?

Maybe, maybe not.  I think that's a discussion to happen outside of the
context of this patch, since the changes I made in the glade files will
be wiped out by deleting the extra widgets if we kill the verification
entries.

> Also: I'm not sure we should do this when *requesting* a passphrase to
> unlock something. It's fine for things like wireless networks or iscsi
> authentication (which are generally shared passphrases) but will make
> people really nervous for things like disks or usernames (i.e. personal
> passphrases).

I'm not convinced, but I'd be willing to listen to what other people
have to say.

> We might need to start our own UI style guide / HIG doc..

I've considered this, yeah.  All the stuff about mnemonic widgets,
spacing, and word wrapping for more verbose languages is pretty
important and easy to forget.  This new signal handler behavior (or for
another example, the iscsi stuff where I check credentials every time an
entry changes) is a good example of something new that everyone needs to
remember.

- Chris


More information about the anaconda-patches mailing list