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
54810 5612 d 11 h pd /branches/R-2-13-branch/ Create R-2-13-branch  
54776 5614 d 17 h ripley /trunk/ improve error message src/main/grep.c  
52965 5786 d 22 h luke /trunk/src/main/ In real_binary use same code structure used to avoid cost of
mod_iterate for handling scalar and equal length args in addition,
etc, for power op as well. Should probably be abstracted into a macro
and be used in more places.		  
52946 5789 d 14 h luke /trunk/src/main/ Applied Radford Neal's patch-vec-arith with some white space editing.  
52938 5791 d 18 h luke /trunk/src/main/ Added inline function R_POW to handle y == 2.0 case inline (part of
dealing with Radford Neal's patch-square)		  
52937 5791 d 19 h luke /trunk/src/main/ Put test for y == 2.0 (squaring) at the top of R_pow (part of
addressing Radford Neal's patch-square).		  
51790 5941 d 3 h ripley /trunk/src/main/ finish tidying up, correct last change  
51781 5941 d 23 h ripley /trunk/src/main/ remove unused static function, add comment  
51702 5949 d 12 h jmc /trunk/ fix for bug 13209  
51657 5952 d 17 h luke /trunk/src/ Modified vectorized bessel functions to allocate one work array per
vectorized call rather than one allocation per element evaluation.		  
51316 5974 d 16 h ripley /trunk/ a few more checks of argument names, and a test suite  
51270 5979 d 0 h ripley /trunk/ separate out primitives in names.c
no longer treat atan() separately
change log{2,10} to avoid double evaluation of first arg.		  
51267 5979 d 4 h ripley /trunk/ check arity and args of one-argument primitives  
51253 5981 d 1 h ripley /trunk/ back out r51251: causes a problem in Rmpfr  
51251 5981 d 3 h ripley /trunk/ avoid double evaluation of args to log2 and log10  
51245 5982 d 1 h ripley /trunk/ document in source files which primitives/.Internal are 'special'
make zero-arg primitives consistently builtin and not special, and check arity
list the specials (and some reasons) in R-ints
update now switch() is primitive		  
49236 6191 d 0 h maechler /trunk/ revert the "double" part of r49062 ["z^n"]  
49076 6199 d 2 h ripley /trunk/src/ replace R_MSG_IA etc by more informative messages  
49062 6200 d 20 h maechler /trunk/ more accurate z^n  
49001 6209 d 1 h ripley /trunk/src/main/ some missed TRACE and SET_TRACE calls  

←Prev12Next→

Show All