randomness_in_r_rdb_rds_databases not very random?

[Chris Lamb]

Hi Vagrant,

> Because of this discrepancy, I created the issue
> "captures_build_path_in_r_rdb_rds_databases" but I also noticed the
> description for "randomness_in_r_rdb_rds_databases" mentions that it is
> the result of a build path:
>
>   Randomness seems to come from using absolute paths in .rd[bs] files.

I think a factor here is the autoclassification script that, as I
understand it, does not differentiate between these two cases.

Now, I did spend significant time on R a couple of months ago, only to
realise that a proper fix (let alone an upstreamable one) is quite
some work away. Which, more relevant to this conversation, meant that
delineating the finer distinctions between R's many issues became less
of a priority. (This is all background and/or explanation for the
status quo, mind you; I would agree it's always better to have better
distinctions.)

> I think most of the fixes mentioned in the description of
> "randomness_in_r_rdb_rds_databases" have been applied.

Interesting. I'm not quite as confident about this as you seem to be,
but glad to be re-assured.


Best wishes,

-- 
      o
    ⬋   ⬊      Chris Lamb
   o     o     reproducible-builds.org 💠
    ⬊   ⬋
      o