File::UStore - UUID based File Storage Module.
use File::UStore; my $store = new File::UStore( path => "/home/shantanu/.teststore", prefix => "prefix_", depth => 5 ); open( my $FH, "foo.pl" ) or die "Unable to open file "; # Add a file in the store my $id = $store->add(*$FH); # Get file handle from uuid. my $handle = $store->get($id); print <$handle>; # Return the filesystem location from a uuid my $location = $store->getPath($id); # Remove a file by its uuid from the store $store->remove("7d4d873e-4bf4-41a5-8696-fd6232f7bdda");
File::UStore is a perl library based on File::HStore to store files on a filesystem using a UUID based randomised storage with folder depth control over storage.File::UStore is a library that allows users to abstract file storage using a UUID based pointer instead of File Hashes to store the file. This is a critical feature for code which requires even duplicate files to get a unique identifier each time they are added to a store. A Hash Storage on the other hand will not allow a file to be duplicated if it is stored multiple time in the same store. This can cause issues in cases where files are deleted regularly as there would be no way of knowing if a second process is still using the file which the first process might be about to delete.
The current version uses UUID Module to generate universally unique identifiers everytime a file is to be stored.
The Module also provides a option to choose the folder depth at which a file is stored. This can be changed from the default value of 3. Increasing depth is advisable if the store might contain a large number of files in your use case. This will help to avoid having a too large number of files in any single folder.
$store = File::UStore->new( path => "/home/shantanu/.teststore", prefix => "prefix_", depth => 5 );
This constructor returns a new
File::UStore object encapsulating a specific store. The path specifies where the UStore is located on the filesystem. If the path is not specified, the path ~/.ustore is used. The $prefix is an extension to specify the prefix appended before unique file name.
my $id = $store->add($filename)
The $filename is the path of the file to be added in the store. The return value is the uuid ($id) of the file stored. From this point on the user will only be able to refer to this file using $id.
Returns undef on error.
The $id is the uuid of the file to be removed from the store.
Returns false on success and undef on error.
Returns the file handle of the file from its uuid. Returns undef on error.
Returns the filesystem location of the file from its uuid. Returns undef on error.
* An Analysis of Compare-by-hash - for reasons why a UUID based storage maybe preferred over hash based solution in certain cases. http://www.usenix.org/events/hotos03/tech/full_papers/henson/henson.pdf
File::HStore is a similar module that provides File Hash based storage. However due to the nature of File Hashing, File::HStore doesn't allow duplicates. If the same file is stored a second time using File::HStore it transparently returns the same hash it had returned last time as an id without adding any new file in storage due to inherent character of hash based storage, while this is useful if a user doesn't want any duplicates occurring in a storage, this apparently trivial difference is risky in the use case where two processes upload a duplicate file to the store and both processes want to do file handling on these files simultaneously, only one of the processes will be able to get a lock(deletion,manipulation etc.) on the file at a time and if the first process deletes the file referred to by its ID, the second process will never know what happened to the file it added. However in circumstances where filename based deduplication is desired you must use L<File::HStore> instead.
Hence to serve this unique use case I wrote this module for a UUID based storage solution which is not hostage to auto de-duping features of HStore. This module returns a unique file id each time a file is uploaded even if its a duplicate of existing file previously uploaded. This allows multiple processes handling data from a common file dump to access the same file. This module also expands on the Hstore to allow the user flexibility of choosing the depth of storage to optimize the performance for the users application. Depth of storage allows users to make the balance between average "number of files in a folder" and folder depth.
Thanks to Alexandre Dulaunoy for the excellent File::HStore module which along with my own special need provided the idea behind this module.
Please report any bugs or feature requests through github at https://github.com/shantanubhadoria/file-ustore/issues. You will be notified automatically of any progress on your issue.
This is open source software. The code repository is available for public review and contribution under the terms of the license.
Shantanu Bhadoria <firstname.lastname@example.org>
This software is copyright (c) 2013 by Shantanu Bhadoria.