[R-pkg-devel] irrecoverable exception occurred during CRAN submission checks

[Daniel Lüdecke]

Hi Dirk,

the package in question is "insight"
(https://cran.r-project.org/package=insight). Since a few days, the test for
the panelr-package fails, because panelr itself fails on many systems
(https://cran.r-project.org/web/checks/check_results_panelr.html). I *think*
this is due to an update of dpylr that causes trouble with panelr, because
both insight and panelr checks were OK the last weeks.

Now win-builder checks for insight are OK, local checks (Win) are OK and
GitHub CI is ok (https://github.com/easystats/insight). 

If you need, I uploaded the source version of insight here:
https://drive.google.com/file/d/1sZ2n3utc27U0DrcFlAsH6NQpKMssIiEg/view?usp=s
haring

The GitHub version only differs in the package version number in the
DESCRIPTION file (which is 0.8.4.1 on GitHub and will be updated once the
update is accepted on CRAN).

I can't imagine any special revisions that might lead to that NOTE in the
incoming checks on Linux.

Hope that helps...

Best
Daniel

-----Ursprüngliche Nachricht-----
Von: Dirk Eddelbuettel <edd using debian.org> 
Gesendet: Donnerstag, 4. Juni 2020 02:01
An: Daniel Lüdecke <d.luedecke using uke.de>
Cc: r-package-devel using r-project.org
Betreff: Re: [R-pkg-devel] irrecoverable exception occurred during CRAN
submission checks


Daniel,

On 3 June 2020 at 10:18, Daniel Lüdecke wrote:
| I just submitted an update of my package to CRAN and it does not pass
| incoming checks. While local checks, win-builder and GitHub CI
| (Mac/Win/Linux) all pass, I get the Note you see below for the checks
during
| submission process. I have no idea why this happens and see this message
for
| the first time. I *assume* it's a temporary issue / false positive, but
want
| to be sure before I reply to the CRAN team.
| 
| Can anyone help how I can address this issue, or if it indeed might be a
| false positive?

What package? A package / repo name would have helped here. Did I miss it?

One of my golden rules for debugging is the old "doctor, doctor, I hurt when
I do this" joke. There is no shame in disabling a test that bites: "if it
hurts, don't do it". The next step then is is to reproduce locally---which
is
still an uphill battle because CRAN, resource contrained as they are, have
not provided their setups in ways that are easily reproducible by us
mortals.

Hth,  Dirk

-- 
http://dirk.eddelbuettel.com | @eddelbuettel | edd using debian.org

--

_____________________________________________________________________

Universitätsklinikum Hamburg-Eppendorf; Körperschaft des öffentlichen Rechts; Gerichtsstand: Hamburg | www.uke.de
Vorstandsmitglieder: Prof. Dr. Burkhard Göke (Vorsitzender), Joachim Prölß, Prof. Dr. Blanche Schwappach-Pignataro, Marya Verdel
_____________________________________________________________________

SAVE PAPER - THINK BEFORE PRINTING