[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bioconductor + R-base transition: next steps and status update


On Mon, Sep 20, 2021 at 04:16:52PM -0500, Dirk Eddelbuettel wrote:
> On 21 September 2021 at 01:58, Nilesh Patra wrote:
> | On 9/19/21 8:16 PM, Dirk Eddelbuettel wrote:
> | > [...]
> | > vegan 2.5.7 is current
> | 
> | I'm not too sure why this one does not work. On a quick look, _seems_ like some float-precision
> | errors. I'm tempted to disable that one and upload.
> | Any other ideas welcome.
> Those things happen sometimes. For my (upstream) package Rcpp I frequently
> run reverse-depends checks against essetially all 2300+ CRAN packages
> depending on it and I currently have three or so that go belly up like
> that. I usually just ignore them (for the purposes of my tests, nothings
> depends upon them). Sometimes it goes aways under rebuilds of all parts.
> The r-base block is now at    "36 days old (needed 5 days)".

Hmmm, did you created a transition bug to coordinate with release team?
This would be pretty helpful to do a transition and could focus our
> If you all could do what it takes to no longer let your packages block it,

We are doing something[1].

> I will not be alone in appreciating availability of r-base 4.1.1 in testing.

Definitely not.  We all want this.  Regarding the packages that are
not up to date I left some notes about blocking reasons inside the
according changelogs.  For instance the changelog of r-cran-processx[2]

══ Failed ══════════════════════════════════════════════════════════════════════
── 1. Failure (test-err.R:252:3): errors from subprocess ───────────────────────
`err` is not an S3 object

── 2. Failure (test-err.R:253:3): errors from subprocess ───────────────────────
err$parent is not an S3 object

── 3. Failure (test-err.R:254:3): errors from subprocess ───────────────────────
is.null(err$parent$trace) is not FALSE

`actual`:   TRUE 
`expected`: FALSE

My strategy was to update all those packages first that will go smoothly
but it seems r-cran-processx is now a blocker for several others and this
should be fixed.  Any hint for this?

Kind regards


[1] http://blends.debian.net/liststats/uploaders_r-pkg.png
[2] https://salsa.debian.org/r-pkg-team/r-cran-processx/-/blob/master/debian/changelog


Reply to: