perl 倒排

rubyish

yige lizi:

DATA: 1 hang => 1 ge doc

1. Hello world pop world
   
2. good hello world
   
3. very good
   
4. Real World Studios
   
5. push pop shift pop pop
   
6. good very very good
   
7. push pop pop insert

复制代码

test:

1. hello world
   
2. HELLO
   
3. kitty
   
4. pop
   
5. Very Good

复制代码

output:

1. [ hello world ]         1, 2, 4
   
2. [ HELLO ]         2, 1
   
3. [ kitty ]         
   
4. [ pop ]         5, 7, 1
   
5. [ Very Good ]         6, 3, 2

复制代码

biru:

1. #!/usr/bin/perl
   
2. 
   
3. use 5.010;
   
4. 
   
5. # init
   
6. 
   
7. my %data;
   
8. 
   
9. while (<DATA>) {
   
10.     add( $., $_ );
    
11. }
    
12. 
    
13. # test
    
14. 
    
15. my @test = ( 'hello world', 'HELLO', 'kitty', 'pop', 'Very Good' );
    
16. 
    
17. for my $test (@test) {
    
18.     my @result = search($test);
    
19.     say "[ $test ]\t ", join ', ', @result;
    
20. }
    
21. 
    
22. # sub
    
23. 
    
24. sub add {
    
25.     my ( $id, $text ) = @_;
    
26.     my @words = map lc, split /\s+/, $text;
    
27.     for my $word (@words) {
    
28.         $data{$word}{$id}++;
    
29.     }
    
30. }
    
31. 
    
32. sub search {
    
33.     my $search = shift;
    
34.     my @words = map lc, split /\s+/, $search;
    
35.     return these( \@words );
    
36. }
    
37. 
    
38. sub these {
    
39.     my @words = @{ +shift };
    
40.     my %good;
    
41.     for my $word (@words) {
    
42.         for my $ID ( keys %{ $data{$word} } ) {
    
43.             $good{$ID}{ID}++;
    
44.             $good{$ID}{COUNT} += $data{$word}{$ID};
    
45.         }
    
46.     }
    
47. 
    
48.     sort {
    
49.              $good{$b}{ID} <=> $good{$a}{ID}
    
50.           || $good{$b}{COUNT} <=> $good{$a}{COUNT}
    
51.     } keys %good;
    
52. }
    
53. 
    
54. __DATA__
    
55. Hello world pop world
    
56. good hello world
    
57. very good
    
58. Real World Studios
    
59. push pop shift pop pop
    
60. good very very good
    
61. push pop pop insert
    

复制代码