Merging Arrays without duplicates

perl_krish has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: Merging Arrays without duplicates by ccn (Vicar) on Aug 18, 2004 at 13:35 UTC
simple merge `@merge = (@A, @B)` for more tricks see perlfaq4 {Data Arrays} upd: To merge without duplicates: `@A = (1,2,3,4,5,6); @B = (1,2,3,4,5,6,7,8,9); @seen{@A} = (); @merged = (@A, grep{!exists $seen{$_}} @B); print @merged;` [download]	[reply] [d/l] [select]
Re^2: Merging Arrays without duplicates by perl_krish (Initiate) on Aug 18, 2004 at 14:06 UTC
yup jus got there (to faq4 i mean) ...... guess now i got to become a perl junkie :D	[reply]
Re^2: Merging Arrays without duplicates by perl_krish (Initiate) on Aug 18, 2004 at 14:31 UTC
hey i am new to this ... could u explain real quick wat lines 3 and 4 do? Thanks Krish	[reply]
Re^3: Merging Arrays without duplicates by ikegami (Patriarch) on Aug 18, 2004 at 14:57 UTC
`@seen{@A} = ();` is the same as `$seen{$A[0]}) = undef; $seen{$A[1]}) = undef; $seen{$A[2]}) = undef; $seen{$A[3]}) = undef; ...` [download] `@merged = (@A, grep{!exists $seen{$_}} @B);` is the same as `@C = grep{!exists $seen{$_}} @B; @merged = (@A, @C);` [download] `@C = grep{!exists $seen{$_}} @B;` @C is assigned the contents of @B not seen in %seen. In other words, @C is assigned the contents of @B not seen in @A.	[reply] [d/l] [select]
Re^4: Merging Arrays without duplicates by perl_krish (Initiate) on Aug 18, 2004 at 19:16 UTC
Re^5: Merging Arrays without duplicates by ikegami (Patriarch) on Aug 18, 2004 at 20:22 UTC
Re: Merging Arrays without duplicates by gellyfish (Monsignor) on Aug 18, 2004 at 13:37 UTC
If you want to perform a merge which removes the duplicates then you might want some thing like: `my %seen; my @out = grep( !$seen{$_}++, @A, @B)` [download] There are other examples in the perlfaq4 /J\	[reply] [d/l]
Re: Merging Arrays without duplicates by rsteinke (Scribe) on Aug 18, 2004 at 13:40 UTC
To remove duplicate elements, use an intermediate hash: `my @out = keys %{{map {($_ => 1)} (@A, @B)}};` [download] The double %{{}} on the outside are necessary, as the inner pair creates a hash reference, and the outer %{} dereferences it. Ron Steinke <rsteinke@w-link.net>	[reply] [d/l]
Re: Merging Arrays without duplicates by gaal (Parson) on Aug 18, 2004 at 13:40 UTC
Do you need to maintain the original order? Is the result supposed to be sorted? If the answer to the first question is "no" or if the answer to the second one is "yes", this will work: `my @merged = sort { $a <=> $b } keys %{ { map { $_ => 1 } } } @A, @B;` If order isn't an issue simply remove the `sort { $a <=> $b }` bit.	[reply] [d/l] [select]
Re: Merging Arrays without duplicates by Mr_Person (Hermit) on Aug 18, 2004 at 15:25 UTC
And just for fun, here's another way to do it using Tie::Array::Unique. `use Tie::Array::Unique; my @a = (0,1,2,3,4,5,6,7); my @b = (3,4,5,6,7,8,9,10); tie my(@merged), 'Tie::Array::Unique', (@a,@b); print join(',', @merged);` [download] Will output `0,1,2,3,4,5,6,7,8,9,10`.	[reply] [d/l] [select]
Re: Merging Arrays without duplicates by bgreenlee (Friar) on Aug 18, 2004 at 13:45 UTC
Sounds suspicously like a homework problem...but anyway, if you don't care about order, just dump them all into a hash. If you do care about order, there are some modules in CPAN that will do the trick: Tie::IxHash, Tie::Hash::Indexed, Tie::LLHash. If want to preserve order and cook it yourself, one way to do it is to make a hash with the data from one of the arrays (probably the larger), then iterate through the other array, pushing onto the first array any elements in the second array that aren't found in the hash. -b	[reply]
Re: Merging Arrays without duplicates by husker (Chaplain) on Aug 18, 2004 at 15:44 UTC
If this isn't homework, then using Set::Array might save you a lot of time.	[reply]