1. 27 Mar, 2022 1 commit
  2. 26 Mar, 2022 3 commits
    • Vincent Goulet's avatar
      Cosmetic · 852d4f32
      Vincent Goulet authored
    • Vincent Goulet's avatar
      Add NEWS entry · dc97160f
      Vincent Goulet authored
    • Vincent Goulet's avatar
      Replace DOUBLE_EPS by DBL_EPSILON · 4cb5d604
      Vincent Goulet authored
      From B.D. Ripley 2022-03-26:
      NEWS for 4.2.0 alpha (and R-devel for months before) has
         • The legacy S-compatibility macros DOUBLE_* in R_ext/Constants.h
           (included by R.h) are deprecated.
      and these packages use them. The plan is to remove them in R-devel
      soon after the release of 4.2.0, so please do so in your package
      before Apr 23.
      The standard C99 constants are those used in that header:
      #define DOUBLE_EPS     DBL_EPSILON
      #define DOUBLE_XMAX    DBL_MAX
      #define DOUBLE_XMIN    DBL_MIN
      defined in <float.h> or for C++, <cfloat>, so please change to those.
      Unfortunately there is no portable way to give a deprecation warning
      in a header so once these are removed your package will fail to
  3. 23 Jan, 2022 3 commits
    • Vincent Goulet's avatar
      Prepare NEWS file for 3.2-1 release · 36189eaf
      Vincent Goulet authored
    • Vincent Goulet's avatar
      Fix misuse of all.equal · 0201cb8e
      Vincent Goulet authored
      Date: Sun, 23 Jan 2022 09:23:34 +0000
      From: Prof Brian Ripley <ripley@stats.ox.ac.uk>
      Subject: CRAN packages misusing all.equal
      This is documented as
            all.equal(target, current, ...)
            ## S3 method for class 'numeric'
            all.equal(target, current,
                      tolerance = sqrt(.Machine$double.eps), scale = NULL,
                      countEQ = FALSE,
                      formatFUN = function(err, what) format(err),
                      ..., check.attributes = TRUE)
         EpiModel MetricsWeighted PLNmodels actuar clustermq estimatr lme4
         nvmix optimall qgcomp rockchalk sensR univariateML updog ursa
      (and others reported separately) call it with three unnamed arguments,
      the first of which is numeric so the third gets matched to 'tolerance'.
        As NEWS for R-devel says
           • all.equal.numeric() gains a sanity check on its tolerance
             argument - calling all.equal(a, b, c) for three numeric vectors
             is a surprisingly common error.
      See the package's CRAN res...
    • Vincent Goulet's avatar
      Move version number to 3.2-1 · 25e36998
      Vincent Goulet authored
  4. 06 Oct, 2021 1 commit
  5. 01 Oct, 2021 1 commit
  6. 30 Sep, 2021 6 commits
  7. 14 Aug, 2021 1 commit
  8. 02 Jun, 2021 12 commits
  9. 01 Jun, 2021 3 commits
  10. 31 May, 2021 6 commits
  11. 30 May, 2021 3 commits