Re: Password leakage avoidance

On 12/24/23 12:22, Tom Lane wrote:
> Joe Conway <mail(at)joeconway(dot)com> writes:
>> Completely unrelated process bikeshedding:
>> I changed the naming scheme I used for the split patch-set this time. I
>> don't know if we have a settled/documented pattern for such naming, but
>> the original pattern which I borrowed from someone else's patches was
>> "vX-NNNN-description.patch".
>
> As far as that goes, that filename pattern is what is generated by
> "git format-patch". I agree that the digit-count choices are a tad
> odd, but they're not so awful as to be worth trying to override.

Ah, knew it was something like that. I am still a curmudgeon doing
things the old way ¯\_(ツ)_/¯

>> The new pattern I picked is "description-vXXX-NN.patch" which fixes all
>> of those issues.
>
> Only if you use the same "description" for all patches of a series,
> which seems kind of not the point. In any case, "git format-patch"
> is considered best practice for a multi-patch series AFAIK, so we
> have to cope with its ideas about how to name the files.
Even if I wanted some differentiating name for the individual patches in
a set, I still like them to be grouped because it is one unit of work
from my perspective.

Oh well, I guess I will get with the program and put every patch-set
into its own directory.

--
Joe Conway
PostgreSQL Contributors Team
RDS Open Source Databases
Amazon Web Services: https://aws.amazon.com