Joris Meys
2018-Nov-07 09:30 UTC
[Rd] Checksums of packages change on CRAN without change in version number
Dear all, I report the same problem again, as one of our sysadmins at the university ran into the same issue again. Every so often they find an in-place update of an R package which almost always amounts to a small change in the DESCRIPTION file. This does cause the checksums to change, and we end up with two packages with the exact same version number but different checksums. To date there hasn't been any satisfactory answer on how to avoid having to update our own configuration builds a week after the initial update due to these stealthy checksum changes. It's also an increasing safety concern for our IT department, as they seriously frown upon software that makes in-place changes. According to them, there's no way they can know whether it is a legitimate change to the package, or the result of some kind of hack. They have to rely on trust, which is a dangerous thing to do in an IT context and which they like to avoid. github report on the issue (this time it's mgcv) : https://github.com/easybuilders/easybuild-easyconfigs/pull/7122#pullrequestreview-172031060 Cheers Joris -- Joris Meys Statistical consultant Department of Data Analysis and Mathematical Modelling Ghent University Coupure Links 653, B-9000 Gent (Belgium) <https://maps.google.com/?q=Coupure+links+653,%C2%A0B-9000+Gent,%C2%A0Belgium&entry=gmail&source=g> ----------- Biowiskundedagen 2017-2018 http://www.biowiskundedagen.ugent.be/ ------------------------------- Disclaimer : http://helpdesk.ugent.be/e-maildisclaimer.php [[alternative HTML version deleted]]
Apparently Analagous Threads
- download.file does not process gz files correctly (truncates them?)
- Date class shows Inf as NA; this confuses the use of is.na()
- truncation/rounding bug with write.csv
- Apparent bug in behavior of formulas with '-' operator for lm
- data.table not available as win binary for R 3.5 yet?