[Rd] Cannot Change Function (PR#14041)

Steven McKinney smckinney at bccrc.ca
Wed Nov 4 22:04:00 CET 2009



> -----Original Message-----
> From: r-devel-bounces at r-project.org [mailto:r-devel-bounces at r-
> project.org] On Behalf Of Stefan Evert
> Sent: Wednesday, November 04, 2009 4:40 AM
> To: michael_karsh at earthlink.net
> Cc: R-bugs at r-project.org; r-devel at stat.math.ethz.ch
> Subject: Re: [Rd] Cannot Change Function (PR#14041)
> 
> What makes you think this is a bug in R?
> 
> > Whenever I try changing a function, it keeps coming up with the same
> > error
> > message.
> >
> > I have the function
> > CN_state_log_sum=function(Tot_log_sum){ #estimate copy number state
> > for the log
>   [...]
> > }
> >
> > When I try to run it in the loop:
> > for (j in 1:length(BB_mean_ref)){	# find copy number states (0 for
> > homozygous
> > deletion, 1 for hemizygous deletion, 2 for neutral, 3 for single
> > amplification,
> > 4 for multiple amplification)
> > 	state_log_sum[j]=CN_state_log_sum(Tot_log_sum[j])
> > 	state_sum_log[j]=CN_state_sum_log(Tot_sum_log[j])
> > }
> > I get the error message:
> > Error in Im(Tot_sum_log) != 0 || Re(Tot_sum_log) <- 2 :
> >  could not find function "||<-"
> 
> The error message indicates that the problem happens in the function
> CN_state_sum_log() rather than CN_state_log_sum(), so it's hardly
> surprising that changing the latter doesn't have any effect.
> 
> My guess: you've go a condition (Re(Tot_sum_log)<-2), which is
> mistaken for the assignment operator "<-" by the R parser. Adding some
> whitespace (Re(Tot_sum_log) < -2) should help both R and human readers
> to make sense of it.

Adding parentheses around the negative number is the only way to keep
such an expression robustly safe given the various whitespace alterations
that happen as such text passes through different parsers, editors,
mailers, proportional pitch font displays (which can fool human readers) ...

( ( Im(Tot_sum_log) != 0 ) || ( Re(Tot_sum_log) < (-2) ) )

Best

Steven McKinney

> 
> Best,
> Stefan
> 
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel



More information about the R-devel mailing list