Hi monks,

I've been developing (FTFOI) a Perl module to manage text files containing complex data structures as an array. It works by first indexing all the data and tie-ing an array to it. Then, you can read/write from/to the file through a tied array. Initially, the "rules of the game" were:

• No database needed: It was designed to read, change and create data files very rapid and easily.
• Fast and memory friendly: Data files can be big. The module uses buffered input/output and the file is not read into memory, but more work is needed! :-) )
• Reusable with different data types: The rationale here is: "If you can index it, you can access it" :-). Only the "sub" that calculates the index is "data specific"

In general it behaves similar to Tie::File except that it manages complex data structures

>Seq1
AGAGAGAGAGAGGA
AGAGAGAGTTCCTT
CATCGCACATCAGA
AGG
>Seq2
GCGTGAGATTGAGG
GAATTGAGAGAGAG
>Seq3
GAGAGCGCGACCCC
TGAGTAGATAGAGA
ACAGATATAGGAGA
AGAG
[download]

#Print the file
print $_.$/ for (@fastafile);

#Reverse the file
@fastafile = reverse @fastafile;

#Change the 2nd secuence
$fastafile[1] = $other_seq;

#Put more sequences
push @fastafile, $other_seq;
[download]

tie @fastafile,'Tie::File',recsep=>"\n>",autochomp=>1;
[download]

As I said before, this module was created "for the fun of it", but I was curious to know if a final version of it could be generally interesting. The idea is too simple, and surely there are modules (probably better than mine) doing the same tasks. I looked at CPAN but, although some other modules can do similar things (Tie::File, DB_File, Bio::DB::Fasta), none of them seem do the same work. Does anybody know how to do the same things with existing modules?

Thanx all

citromatik