> -----Original Message-----
> From: Rob Dixon [mailto:[EMAIL PROTECTED]
> Sent: Monday, January 14, 2008 8:59 PM
> To: [email protected]
> Cc: Kevin Viel
> Subject: Re: references to complex data structures
>
> Kevin Viel wrote:
> > Consider the following anonymous array of anonymous arrays:
> >
> > my $a = [ [ 00 , 01 ]
> > , [ 10 , 11 ]
> > ] ;
> >
> > I can print the first element of the first anonymous array:
> >
> > print "$$a[ 0 ][ 0 ]\n" ;
> >
> > or, equivalently, I can explicit use bracers to dereference it:
> >
> > print "${$a}[ 0 ][ 0 ]\n" ;
>
> OK
>
> But they're 'braces' :)
>
> > Should I not need two pairs of bracers?
>
> The 'missing' pair of braces is because you're using both of
> the two different syntaxes for indexing a referenced array.
> The first element of the outer array is either
>
> ${$a}[0] (which can be written $$a[0] without ambiguity)
>
> or
>
> $a->[0]
>
> These are references to the first inner array, which can also
> be indexed in either way. So we have four ways of reaching
> the first element of the first inner array:
>
> ${${$a}[0]}[0]
>
> ${$a}[0]->[0]
>
> ${$a->[0]}[0]
>
> $a->[0]->[0]
>
> In addition, Perl allows us to remove the arrow operator
> between pairs of closing and opening brackets or braces - ][
> or }{, so the second and last options can be written
>
> ${$a}[0][0]
>
> $a->[0][0]
>
> and so your code turns out to be the same as the first of
> these two. If you use the same syntax for indexing in both
> cases then the second pair of braces reappears. It is also
> the same as your final example below.
>
> > $a is a reference to the anonymous array. The elements of the
> > anonymous arrays are references to anonymous arrays, correct?
>
> They are references to arrays, yes. They may be named arrays
> or anonymous ones.
>
> > The following seems to achieve this result:
> >
> > print "${${$a}[ 0 ]}[ 0 ]\n" ;
> >
> > Is the outmost pair of bracers with the appropriate symbol
> ($, @, %)
> > the default? If so, how does perl select the correct symbol?
>
> I'm not sure what you mean here: you have explicit dollar
> signs in this code for both dereferences. Perl has no default
> way of handling references, but it will complain if you try
> to use a reference to one type of data as something different.
>
> > I realize that it is seemingly moot, but it may help my
> understanding
> > of more complex structures, like hash of arrays or hash of hases.
>
> Don't get to like them to the extent that you use them when
> your data isn't shaped that way. They are a way of expressing
> hierarchical data only.
Thanks for another informative post.
My data are hierarchical. I have 22 files, one for each autosome (non-sex
chromosome, sex chromosomes being the X and Y-apologies is this is
unnecessary to state). In each file I have markers and their locations.
The first thing I want to do is to be sure that that a marker is unique to a
chromosome:
if ( exists $marker_hash{ $marker } ) {
print "Collision: Chromosome: $Chr Marker: $marker\n" ;
}
else{
$marker_hash{ $marker } = $chr ;
}
However, I would also like to do two things: collect the chromosomes on
which a markers exists and create a ordered database table that reports the
chromosome , marker , and location.
The outer most object would be a hash, the keys of which are the markers.
The values of this hash could be a hash with keys being chromosome and the
values being an array of locations.
Coding aside, through which I still must suffer, does a complex data
structure seem appropriate to this problem or should I be considerng another
approach?
Kind regards,
Kevin
Kevin Viel, PhD
Post-doctoral fellow
Department of Genetics
Southwest Foundation for Biomedical Research
San Antonio, TX 78227
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
http://learn.perl.org/
