[ about | pics | projects | repos | weblog ]

[ ~derf :: projects :: App::Hashl v1.00 ]

hashl creates a database with partial file hashes. Using it, you can for example leech files off slow FTPs withouth wasting bandwidth on duplicates you already have; or you can keep a list of files on an external database and always tell people what you have and what you don't – and find out which files in a given directory are not on the external disk.

hashl only takes the first few MiB of every file into account, giving it a huge speed advantage at the cost of possible (though highly improbable) false positives. Do not use it for file integrity checks.

For further details on these usecases, please see hashl(1).


App::Hashl v1.00

App-Hashl-1.00.tar.gz (signature)

  • Add "hashl list regex" to filter listed files by name
  • Do not fail when encountering zero-length files

App::Hashl v0.2

App-Hashl-0.2.tar.gz (signature)

  • The internal representation of ignored hashes was changed
  • hashl update no longer re-adds ignored (but existing) files
  • hashl update tries to save its state when being interrupted (sigint/sigterm)
  • add hashl -f update - update db, unignore and add any ignored files in the scanned directory

hashl v0.1

hashl-0.1.tar.gz (signature)

  • Initial release
  • Create/Update hash database
  • Find/Copy files not in database
  • Ignore files (will be marked as known, but can be removed from the file system)