Beefy Boxes and Bandwidth Generously Provided by pair Networks
No such thing as a small change
 
PerlMonks  

Re^2: Sorting a "tuple" by the second value

by thmsdrew (Scribe)
on Apr 09, 2012 at 04:53 UTC ( #964086=note: print w/ replies, xml ) Need Help??


in reply to Re: Sorting a "tuple" by the second value
in thread Sorting a "tuple" by the second value

Thanks for that link, it was helpful for sure. Here's how I did it. The first subroutine gets the tuples from the file, splits them into an array, and writes them all into one array.

sub get_links { my @links; open(my $fh, "<", "links.alpha.sorted.25sample") or die "cannot open < links.alpha.sorted.25sample: $!"; while(<$fh>) { chomp; my $tuple; @$tuple = split(/\s+/, $_); push(@links, $tuple); } return @links; }

Then I sort by the second value and write it to another file:

sub sort_and_store { my @links = get_links(); my @sorted_links = sort { $a->[1] cmp $b->[1] } @links; open(my $fh, ">", "sorted.by.destination") or die "cannot open > sorted.by.destination: $!"; foreach my $tuple (@sorted_links) { print $fh "@$tuple\n"; } }

I've never used the @$blah variable type before, and I couldn't find any information about it. I'm guessing that's how you refer to the arrays in an array of arrays?


Comment on Re^2: Sorting a "tuple" by the second value
Select or Download Code
Re^3: Sorting a "tuple" by the second value
by Anonymous Monk on Apr 09, 2012 at 08:13 UTC

    I've never used the @$blah variable type before, and I couldn't find any information about it. I'm guessing that's how you refer to the arrays in an array of arrays?

    Yes, that is one way to do it, known as dereferenceing because $blah is a reference, see references quick reference

    Its nice you picked up a new trick, but I would still use sort :)

    $ cat fafafile 3 6 1 1 2 0 9 1 9 4 4 5 7 3 2 0 5 4 6 7 6 2 9 8 $ sort --dictionary-order fafafile 0 5 4 1 2 0 2 9 8 3 6 1 4 4 5 6 7 6 7 3 2 9 1 9 $ sort --dictionary-order --key=2,2 fafafile 9 1 9 1 2 0 7 3 2 4 4 5 0 5 4 3 6 1 6 7 6 2 9 8

      Wow, that is a lot easier than my way. It takes longer but I also had to take the time to write the program in the first place! Thank you kind sir!

Re^3: Sorting a "tuple" by the second value
by johngg (Abbot) on Apr 09, 2012 at 12:22 UTC

    You could combine the steps in your get_links() and sort_and_store() subroutines into one process.

    knoppix@Microknoppix:~$ cat spw964070.in abc peter def jack ghi zak jkl ben mno mick pqr alan knoppix@Microknoppix:~$ perl -Mstrict -wE ' > open my $inFH, q{<}, q{spw964070.in} > or die qq{open: < spw964070.in: $!\n}; > open my $outFH, q{>}, q{spw964070.out} > or die qq{open: > spw964070.out: $!\n}; > > print $outFH > map { $_->[ 0 ] } > sort { $a->[ 2 ] cmp $b->[ 2 ] } > map { [ $_, split ] } > <$inFH>; > > close $inFH > or die qq{close: < spw964070.in: $!\n}; > close $outFH > or die qq{close: > spw964070.out: $!\n};' knoppix@Microknoppix:~$ cat spw964070.out pqr alan jkl ben def jack mno mick abc peter ghi zak knoppix@Microknoppix:~$

    Sticking to your subroutines, it would be more efficient for get_links() to return a reference to @info than the huge array itself, like so:-

    ... return \ @links; } sub sort_and_store { my $refToLinks = get_links(); my @sorted_links = sort { $a->[1] cmp $b->[1] } @$refToLinks; ...

    I hope these points are helpful.

    Cheers,

    JohnGG

      Definitely helpful. I need to learn more about referencing and when it's best to use. Thanks!

        “Referencing” is a drop-dead simple idea, actually:   a reference is “a tiny thing” that refers to something else in memory.   That thing which is referred-to may be big or small, and there may be just one reference to it or a great many.   But each reference is very small and cheap.   So, for example, if you have in-memory a list of “big records,” you can swap and manipulate and what-have-you those records just by playing around with lists (or arrays, or hashes) of references to them:   nothing actually “moves.”   This lets you build up arbitrarily-complex data structures very easily, and to work with them very efficiently.   (Want to sort the same list of records by three separate keys at once, all in-memory, keeping all three lists, without making duplicate copies of anything?   Feel Free.™)   It’s one of the hallmarks of Perl and one of the compelling reasons for using this language.

Log In?
Username:
Password:

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://964086]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this? | Other CB clients
Other Users?
Others meditating upon the Monastery: (10)
As of 2014-07-25 21:35 GMT
Sections?
Information?
Find Nodes?
Leftovers?
    Voting Booth?

    My favorite superfluous repetitious redundant duplicative phrase is:









    Results (175 votes), past polls