The R Project SVN R

Rev

Show changed files | Details | Compare with Previous | Blame | RSS feed

Filtering Options

Rev Age Author Path Log message Diff
84772 1095 d 6 h maechler /branches/R-OBJSXP/ 'Create R-OBJSXP branch - new OBJSXP type "==" S4SXP = "object" in R'  
84525 1144 d 5 h maechler /trunk/tests/ tightening checks for ks.test() ~~ related to r84320 etc  
83107 1379 d 17 h smeyer /trunk/ add regtest from PR#979  
82037 1579 d 20 h smeyer /trunk/ facilitate basic tests on Windows  
81441 1665 d 4 h ripley /trunk/tests/ use factor() rather than structure()  
81437 1665 d 18 h ripley /trunk/ modernize some uses of structure()  
81418 1669 d 15 h smeyer /trunk/tests/ typo (invalid row.names)  
80378 1889 d 3 h maechler /trunk/tests/ adapt to _R_CHECK_MATRIX_DATA_=TRUE  
80296 1900 d 5 h ripley /trunk/ reduce chatter when checking without recommended packages  
78541 2257 d 5 h maechler /trunk/ type.convert(*, as.is): warning if not specified; in line with R 4.0.0`s adaption to read.table(.., stringsAsFactors = .) defaulting to FALSE  
78051 2315 d 18 h maechler /trunk/tests/ no 'options(stringsAsFactors = TRUE)' anymore  
76160 2708 d 14 h luke /trunk/ New default algorithm for R_unif_rand, used in sample, to resolve PR
17494. Includes adding sample.kind to RNGkind to allow reverting to
the old algorithm for reproducing previous results. Thanks to Duncan
Murdoch for contributing the patch.		  
75767 2791 d 5 h ripley /trunk/ clean up comments, mainly about HP-UX  
75334 2866 d 23 h maechler /trunk/ fix / enhance influence.lm() etc for "mlm" incl Fortran, C  
75313 2872 d 5 h maechler /trunk/tests/ another && (length > 1) fix  
75127 2904 d 23 h hornik /trunk/ Have S3 methods lookup by default look for the S3 registry in the topenv
of the generic.		  
72984 3285 d 20 h maechler /trunk/tests/ "logical cosmetic"  
72133 3457 d 0 h kalibera /trunk/ Make test outputs a bit more deterministic by initializing the random seed
with a constant.		  
72021 3471 d 3 h kalibera /trunk/ Let NA/NaN propagate automatically in cumsum as they do in cumprod. This
improves consistency and prevents NaNs in cumsum from becoming NAs
unnecessarily. Add tests ensuring NAs are preserved.		  
71484 3577 d 4 h maechler /trunk/tests/ smooth.spline() checks are all in ../src/library/stats/tests/smooth.spline.R  

←Prev12Next→

Show All