On Fri, May 15, 2020 at 02:36:02PM -0400, Konstantin Ryabitsev wrote:
> On Fri, May 15, 2020 at 07:19:18PM +0100, Mark Brown wrote:

> > a mailbox is generated (incomplete due to issues with vger spam
> > filtering) but the fetched patches are not recorded in the b4 ty
> > database (as showing by b4 ty -l).  This happens even if only patches
> > that are present on the list are requested on the command line.

> Right, this was intentional due to trying to be over-cautious. If we 
> don't have the complete thread, then we probably shouldn't be tracking 
> if this incomplete thread was applied or not -- at least that was my 
> thinking. What is your thoughts on that?

I think that's a sensible decision and wasn't particularly surprised by
that bit of it.  In fact what might be handy would be an option to not
output the partial mailbox if we can't find some of the patches - that'd
help spot problems early on in the process.

> Of course, if we're specifically selecting a subset of that thread, we 
> shouldn't be applying that logic. I'll see if I can add that caveat in.

Right, I'd have expected it to only apply to whatever was explicitly
asked for.