davido has asked for the wisdom of the Perl Monks concerning the following question:
I was looking at the code for Scalar::Util, and in specific at the source for the looks_like_number() function. There's something strange in the code that prompts me to ask the question of 'why'.
Here is the source for the looks_like_number() function, pasted from Scalar::Util:
sub looks_like_number { local $_ = shift; # checks from perlfaq4 return $] < 5.009002 unless defined; return 1 if (/^[+-]?\d+$/); # is a +/- integer return 1 if (/^([+-]?)(?=\d|\.\d)\d*(\.\d*)?([Ee]([+-]?\d+))?$/); # a C float return 1 if ( $] >= 5.008 and /^(Inf(inity)?|NaN)$/i) or ($] >= 5.006001 and /^Inf$/i ); 0; }
The line I'm questioning is:
return $] < 5.009002 unless defined;
As I understand from reading perlvar, $] contains the version + patchlevel / 1000 of the perl interpreter. The question is why is Scalar::Util's looks_like_number() written to return truth if its parameter is undefined and the version is less than 5.009002? This would mean that the looks_like_number() function will errantly report an undefined value as looking like a number if you're using a sufficiently old version of Perl5. Is this really what is intended? Or am I missing something simple?
Dave
|
|---|
| Replies are listed 'Best First'. | |
|---|---|
|
Re: Scalar::Util 'looks_like_number()' code's $] comparison
by ysth (Canon) on Dec 15, 2004 at 23:55 UTC | |
|
Re: Scalar::Util 'looks_like_number()' code's $] comparison
by eieio (Pilgrim) on Dec 15, 2004 at 23:54 UTC |