indexing flat files

[Siva Sai]

Hi,

I have a file (about 3-4 MB ) with about 20000 records
in it. I need to create an index on a particular field
(integer ) in each record. But there might be multiple
records matching a particular field value. My queries
are something like,
"Get all records with the A < field value < B"

This looks like an ideal candidate for a DBMS solution
like MySQL but I do not have it. Is there any way I
could do this efficiently in Perl. Will the DB_File
module be usefull( I know very little about this ).
Can anyone please help?

Thnx.

Siva.

__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
[url]http://sitebuilder.yahoo.com[/url]

[Awards]

Hi,

this is just an Idea and maybe it is the worst idea for what you want. But I
'll go for it :-)

What I would do-- I have an Index(sorted)
Then the main file with the 20,000 records would be a DBM just a tied or
dynamic or a hashed text.
using the module
SDBM_File, POSIX modules
And the index would be the address or Key of the record in a main file.
like in the index sorted by integers
you would have
YOUR INTEGER | ADDRESS or key

where Address would be the row in the main file.
when you create the 'tied' file, you can have a counter.
foreach $line(@data){
$c++;
($int, $othervalue)=split(/\|/,$line);
$dbm{$c}= $line;
push @tobegotoindex, "$int|$c\n";
}

just an idea.
That is what I used to do with Visual Basic.

awards