NAME - binary search of variable length.


        use File::BetweenTree;

        # Object interface
        $bt = File::BetweenTree->new("log_file") or
                        die "Can't open file: $!";

        # .. Gets between a and b
        $result_array_ref = $bt->search($a, $b);

        # .. or only a
        $result_array_ref = $bt->search($a);

        # .. process monitor
        print $bt->mon;


This module tracks the data instantly from "Sorted file of variable length" with a binary search. It is simple to use, memory efficient and search instantly from the files of 100 million.

You can choose to search in ascending or descending order.Further Can set the offset and the number of results. you can set the input record separator string on a per file basis.


These are the methods in File::BetweenTree' object interface:

new($file, [ $rec_sep ]);

new takes as arguments a filename, an optional record separator. set the default record separator according to this OS.

search is able to find all of the intermediate element "two elements".

        $a, $b, [mode], [limit], [offset], [order_by], [col_sep], [col_num]

look for the data between <$a> and <$b>. <$a>, the maximum and minimum values are <$b>.

It can be the same value <$b> a <$a>.

The return value is an array reference. If you can not find a match, it returns a NULL data. Nearby value is returned to the second. Become undef if the minimum value of the one front does not exist.

        @{$result_array_ref}[0] => NULL
        @{$result_array_ref}[1] => ..Minimum value near

mode is optional; the default is '0' Search target set number or a text string. Settings are required to appropriate an accurate result.

        0 : number string search dedicated
        1 : text string search dedicated.

limit is optional; the default is '1000'

        You can use C<limit> to restrict the scope of the C<search>.

offset is optional; the default is '0'

        This specifies the offset of the first row to return

order_by is optional; the default is 'ASC' Select a search in ascending or descending order

        ASC  : ascending
        DESC : descending

col_sep is optional; the default is ',' You can specify a separator to divide the line.

        example: value_0:value_1:value_2 => ":"

col_num is optional; the default is '0' Order of the sequence obtained by dividing the line.

        example: value_0,value_1,value2 => Value_1 and is "1"

For example:


        my $result_array_ref = $bt->search(
           'lemon', # mininum data
           'orang', # maximum data
           1,       # mode is text string
           10,      # result 10 limit
           0,       # offset 0
           'ASC',   # order_by: ascending
           ':',     # separator to divide the line
           1,       # data of the second from the left
        print "result = " . join("\n", @{$result_array_ref});

        result = 5:lemon 6:lemon 7:lemon 8:orange 9:orange.


You can monitor the state of the processing.

        print $bt->mon;

        file_read:1 roughly_start_addr:420200 search_length:2285

        <=|<- 37415913 18707956
        <=|<- 18707957 9353978
        <=|<- 9353979 4676989
        <=|<- 4676990 2338494
        <=|<- 2338496 1169247
        <=|<- 1169249 584623
        <=|<- 584626 292311
        =>|-> 292315 146155 | min
        <=|<- 438470 73077
        =>|-> 365393 36538 | min
        =>|-> 401931 18269 | min
        =>|-> 420200 9134 | min
        <=|<- 429334 4567
        <=|<- 424767 2283
        <=|   421342 1141 | max

Interpretation above

<- : $a Move to low addr the current pointer. =item -> : $a Move to high addr the current pointer. =item <= : $b Move to low addr the current pointer. =item => : $b Move to high addr the current pointer.
Number of first : current pointer addr. =item Number of second: Amount of movement of the pointer.
| min : Addr with a minimum value determined for the time being. =item | max : Addr with a maximum value determined for the time being.


Mitsuru Yasuda,


Copyright (C) 2013 by Mitsuru Yasuda &

This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself, either Perl version 5.14.2 or, at your option, any later version of Perl 5 you may have available.

2 POD Errors

The following errors were encountered while parsing the POD:

Around line 426:

'=item' outside of any '=over'

Around line 438:

You forgot a '=back' before '=head1'