R-alpha: Second patch for 0.50-a1: cut(.) fails ..

[Martin Maechler]

Thank you for these patches, especially the new color scheme!


``As always'' (;-)  
some remarks:

1)
 	     cut(rep(1,5),4)
>>>> Error: character variables must be duplicated in .C/.Fortran

  Which seems the old '*dup =0' problem, but is NOT (identical at least)
	  [Yes, I made sure that the above was applied dotcode.c]

  This seems only for  cut(...), not for 
	  approx(.),  spline(.) or lowess(.) 
  which also use .C(...).

  Further note that cut(.) had no problems in 0.50-Alpha-2 


2) I think that
	given the definition in	src/include/Fortran.h ,
	src/appl/cpoly.c  should rather use  POW_DI(.)  instead of  pow_di(.).

  It probably won't matter for most (all?) installations,
  since it only becomes important when

#ifdef NONSTANDARD_F77LIBS  
				is true


3) Important(IMHO) bugs [besides '1)' above] remaining:
	
	i.  browser()  ##	-> seg.fault

	ii. options(digits ..) still ``fails''
	    when used in non-global environments	(see my earlier posts)
	
--------
Martin
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-