On 13 Jan 2003, Philip Martin wrote:
> Karl Fogel <kfogel@newton.ch.collab.net> writes:
>
> > Philip Martin <philip@codematters.co.uk> writes:
> > > If the use cases above are really what you want then I think you
> > > should change the store-password default and leave --no-auth-cache as
> > > it is
> > >
> > > Case 1: the default, store-password = no, doesn't have --no-auth-cache
> > > username is stored
> > > password is not stored
> > >
> > > Case 2: store-password = yes, doesn't have --no-auth-cache
> > > username is stored
> > > password is stored
> > >
> > > Case 3: --no-auth-cache, store-password is irrelevant
> > > username is not stored
> > > password is not stored
> >
> > There has to be a way to trigger password storing from the command
> > line, though. It's annoying to have to edit a config file to get this
> > behavior on the fly.
>
> Agreed, although that's a new feature we don't currently have.
>
> When it comes to caching there are three outcomes as shown above (it
> doesn't make much sense to cache the password without the username).
> So far, there seems to be general agreement that the password should
> not be cached by default, and also that there should be a command line
> option for "one shot" password caching. I think we need to retain
> --no-auth-cache as a "one shot" flag to disable all caching. I don't
> know whether the username should be cached by default, but if it isn't
> then we need an option for "one shot" username only caching.
What is the argument for not caching the username?
Ryan
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Mon Jan 13 23:45:02 2003